Uploaded by PullBird

bog

advertisement
MATHEMATICS
FOR ECONOMICS AND BUSINESS
IAN JACQUES
NINTH EDITION
@ChristofVanDerWalt
MyLab Math
MyLab Math provides the practice, instruction and personalised learning path that
students need to master concepts and develop critical numeracy skills.
With over 5 million global student registrations, MyLab Math combines proven results
and engaging experiences to grow student confidence and improve their understanding
of mathematics.
Designed for lecturers and students
MyLab Math provides lecturers with a rich and flexible set of course materials, along
with course-management tools that make it easy to deliver all or a portion of their course
online. It provides students with a personalised interactive learning environment, where
they can learn at their own pace and measure their progress.
Trusted partner
MyLab Math lets you teach your course your way. Ideal as a self-study resource for students
who need extra help, or you can take advantage of the advanced customisation options
and create your own course. MyLab Math allows you to quickly and easily create tests
to assess your students’ skills from basic numeracy right up to undergraduate calculus.
‘Each student received a study plan based on his/her exam result and
practised the required area using MyLab Math. I can now monitor each
student’s progress on a weekly basis . . . MyLab Math also made
a significant saving on the workload for me as the lecturer.’
Dr Shadi Ostavari, University of Greenwich
To find out more, visit:
www.pearson.com/mylab/math-global
This page intentionally left blank
MATHEMATICS
FOR ECONOMICS AND BUSINESS
At Pearson, we have a simple mission: to help people
make more of their lives through learning.
We combine innovative learning technology with trusted
content and educational expertise to provide engaging
and effective learning experiences that serve people
wherever and whenever they are learning.
From classroom to boardroom, our curriculum materials, digital
learning tools and testing programmes help to educate millions
of people worldwide – more than any other private enterprise.
Every day our work helps learning flourish, and
wherever learning flourishes, so do people.
To learn more, please visit us at www.pearson.com/uk
MATHEMATICS
FOR ECONOMICS AND BUSINESS
IAN JACQUES
NINTH EDITION
Harlow, England • London • New York • Boston • San Francisco • Toronto • Sydney • Dubai • Singapore • Hong Kong
Tokyo • Seoul • Taipei • New Delhi • Cape Town • São Paulo • Mexico City • Madrid • Amsterdam • Munich • Paris • Milan
PEARSON EDUCATION LIMITED
Kao Two
Kao Park
Harlow CM17 9NA
United Kingdom
Tel: +44 (0)1279 623623
Web: www.pearson.com/uk
First published 1991 (print)
Second edition published 1994 (print)
Third edition published 1999 (print)
Fourth edition published 2003 (print)
Fifth edition published 2006 (print)
Sixth edition published 2009 (print)
Seventh edition published 2013 (print and electronic)
Eighth edition published 2015 (print and electronic)
Ninth edition published 2018 (print and electronic)
© Addison-Wesley Publishers Ltd 1991, 1994 (print)
© Pearson Education Limited 1999, 2009 (print)
© Pearson Education Limited 2013, 2015, 2018 (print and electronic)
The right of Ian Jacques to be identified as author of this work has been asserted by him in accordance with the Copyright,
Designs and Patents Act 1988.
The print publication is protected by copyright. Prior to any prohibited reproduction, storage in a retrieval system, distribution
or transmission in any form or by any means, electronic, mechanical, recording or otherwise, permission should be obtained
from the publisher or, where applicable, a licence permitting restricted copying in the United Kingdom should be obtained
from the Copyright Licensing Agency Ltd, Barnard’s Inn, 86 Fetter Lane, London EC4A 1EN.
The ePublication is protected by copyright and must not be copied, reproduced, transferred, distributed, leased, licensed or
publicly performed or used in any way except as specifically permitted in writing by the publishers, as allowed under the terms
and conditions under which it was purchased, or as strictly permitted by applicable copyright law. Any unauthorised distribution
or use of this text may be a direct infringement of the author’s and the publisher’s rights and those responsible may be liable in
law accordingly.
All trademarks used herein are the property of their respective owners. The use of any trademark in this text does not vest
in the author or publisher any trademark ownership rights in such trademarks, nor does the use of such trademarks imply any
affiliation with or endorsement of this book by such owners.
Pearson Education is not responsible for the content of third-party internet sites.
ISBN: 978-1-292-19166-9 (print)
978-1-292-19170-6 (PDF)
978-1-292-19171-3 (ePub)
British Library Cataloguing-in-Publication Data
A catalogue record for the print edition is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Jacques, Ian, 1957– author.
Title: Mathematics for economics and business / Ian Jacques.
Description: Ninth edition. | Harlow, United Kingdom : Pearson Education,
[2018] | Includes bibliographical references and index.
Identifiers: LCCN 2017049617| ISBN 9781292191669 (print) | ISBN 9781292191706
(pdf) | ISBN 9781292191713 (epub)
Subjects: LCSH: Economics, Mathematical. | Business mathematics.
Classification: LCC HB135 .J32 2018 | DDC 512.024/33—dc23
LC record available at https://lccn.loc.gov/2017049617
10 9 8 7 6 5 4 3 2 1
22 21 20 19 18
Front cover image © Getty Images
Print edition typeset in 10/12.5pt Sabon MT Pro by iEnergizer Aptara®, Ltd.
Printed in Slovakia by Neografia
NOTE THAT ANY PAGE CROSS REFERENCES REFER TO THE PRINT EDITION
To Victoria, Lewis and Celia
Contents
Preface
xi
INTRODUCTION: Getting Started
1
Notes for students: how to use this text
Chapter 1 Linear Equations
1
5
1.1
Introduction to algebra
1.1.1 Negative numbers
1.1.2 Expressions
1.1.3 Brackets
Key Terms
Exercise 1.1
Exercise 1.1*
6
7
9
12
17
18
20
1.2
Further algebra
1.2.1 Fractions
1.2.2 Equations
1.2.3 Inequalities
Key Terms
Exercise 1.2
Exercise 1.2*
22
22
29
33
36
36
38
1.3
Graphs of linear equations
Key Terms
Exercise 1.3
Exercise 1.3*
40
51
52
53
1.4
Algebraic solution of simultaneous linear equations
Key Term
Exercise 1.4
Exercise 1.4*
55
65
65
66
1.5
Supply and demand analysis
Key Terms
Exercise 1.5
Exercise 1.5*
67
80
80
82
1.6
Transposition of formulae
Key Terms
Exercise 1.6
Exercise 1.6*
84
91
91
92
1.7
National income determination
Key Terms
Exercise 1.7
Exercise 1.7*
Formal mathematics
Multiple choice questions
Examination questions
93
105
105
106
109
112
116
Contents vii
Chapter 2 Non-linear Equations
Quadratic functions
Key Terms
Exercise 2.1
Exercise 2.1*
2.2 Revenue, cost and profit
Key Terms
Exercise 2.2
Exercise 2.2*
2.3 Indices and logarithms
2.3.1 Index notation
2.3.2 Rules of indices
2.3.3 Logarithms
2.3.4 Summary
Key Terms
Exercise 2.3
Exercise 2.3*
2.4 The exponential and natural logarithm functions
Key Terms
Exercise 2.4
Exercise 2.4*
Formal mathematics
Multiple choice questions
Examination questions
2.1
Chapter 3 Mathematics of Finance
Percentages
3.1.1 Index numbers
3.1.2 Inflation
Key Terms
Exercise 3.1
Exercise 3.1*
3.2 Compound interest
Key Terms
Exercise 3.2
Exercise 3.2*
3.3 Geometric series
Key Terms
Exercise 3.3
Exercise 3.3*
3.4 Investment appraisal
Key Terms
Exercise 3.4
Exercise 3.4*
Formal mathematics
Multiple choice questions
Examination questions
3.1
Chapter 4 Differentiation
4.1
The derivative of a function
Key Terms
Exercise 4.1
Exercise 4.1*
121
122
136
137
138
140
148
148
150
151
151
155
161
167
168
168
170
172
182
182
183
186
189
193
197
198
204
208
210
210
213
216
226
226
228
230
238
238
239
241
253
253
255
257
259
263
267
268
277
277
278
viii Contents
4.2
4.3
4.4
4.5
4.6
4.7
4.8
Rules of differentiation
Rule 1 The constant rule
Rule 2 The sum rule
Rule 3 The difference rule
Key Terms
Exercise 4.2
Exercise 4.2*
Marginal functions
4.3.1 Revenue and cost
4.3.2 Production
4.3.3 Consumption and savings
Key Terms
Exercise 4.3
Exercise 4.3*
Further rules of differentiation
Rule 4 The chain rule
Rule 5 The product rule
Rule 6 The quotient rule
Exercise 4.4
Exercise 4.4*
Elasticity
Key Terms
Exercise 4.5
Exercise 4.5*
Optimisation of economic functions
Key Terms
Exercise 4.6
Exercise 4.6*
Further optimisation of economic functions
Key Term
Exercise 4.7*
The derivative of the exponential and natural logarithm functions
Exercise 4.8
Exercise 4.8*
Formal mathematics
Multiple choice questions
Examination questions
Chapter 5 Partial Differentiation
5.1
5.2
5.3
279
279
280
281
286
286
288
290
290
297
299
301
301
302
304
305
307
310
312
313
314
326
326
327
329
345
345
347
348
359
359
361
370
371
373
376
382
389
Functions of several variables
Key Terms
Exercise 5.1
Exercise 5.1*
Partial elasticity and marginal functions
5.2.1 Elasticity of demand
5.2.2 Utility
5.2.3 Production
Key Terms
Exercise 5.2
Exercise 5.2*
390
400
401
402
404
404
407
413
415
416
418
Comparative statics
Key Terms
Exercise 5.3*
420
429
429
Contents ix
5.4
Unconstrained optimisation
Key Terms
Exercise 5.4
Exercise 5.4*
433
444
444
445
5.5
Constrained optimisation
Key Terms
Exercise 5.5
Exercise 5.5*
447
456
457
458
5.6
Lagrange multipliers
Key Terms
Exercise 5.6
Exercise 5.6*
460
468
469
470
Formal mathematics
Multiple choice questions
Examination questions
Chapter 6 Integration
472
474
477
483
6.1
Indefinite integration
Key Terms
Exercise 6.1
Exercise 6.1*
484
495
496
497
6.2
Definite integration
6.2.1 Consumer’s surplus
6.2.2 Producer’s surplus
6.2.3 Investment flow
6.2.4 Discounting
Key Terms
Exercise 6.2
Exercise 6.2*
499
503
504
506
508
509
509
510
Formal mathematics
Multiple choice questions
Examination questions
Chapter 7 Matrices
513
515
518
523
7.1
Basic matrix operations
7.1.1 Transposition
7.1.2Addition and subtraction
7.1.3 Scalar multiplication
7.1.4 Matrix multiplication
7.1.5 Summary
Key Terms
Exercise 7.1
Exercise 7.1*
524
526
527
530
531
539
539
540
542
7.2
Matrix inversion
Key Terms
Exercise 7.2
Exercise 7.2*
545
560
560
561
7.3
Cramer’s rule
Key Term
Exercise 7.3
Exercise 7.3*
564
572
572
573
x Contents
Formal mathematics
Multiple choice questions
Examination questions
Chapter 8 Linear Programming
576
577
581
585
8.1
Graphical solution of linear programming problems
Key Terms
Exercise 8.1
Exercise 8.1*
586
600
601
602
8.2
Applications of linear programming
Key Terms
Exercise 8.2
Exercise 8.2*
604
612
612
614
Formal mathematics
Multiple choice questions
Examination questions
Chapter 9 Dynamics
617
618
623
627
9.1
Difference equations
9.1.1 National income determination
9.1.2 Supply and demand analysis
Key Terms
Exercise 9.1
Exercise 9.1*
628
634
636
639
639
640
9.2
Differential equations
9.2.1 National income determination
9.2.2 Supply and demand analysis
Key Terms
Exercise 9.2
Exercise 9.2*
643
649
651
653
654
655
Formal mathematics
Multiple choice questions
Examination questions
Answers to Problems
658
659
662
664
Chapter 1
664
Chapter 2
674
Chapter 3
683
Chapter 4
687
Chapter 5
698
Chapter 6
705
Chapter 7
709
Chapter 8
715
Chapter 9
719
Glossary
723
Index
730
Preface
This text is intended primarily for students on economics, business studies and management
courses. It assumes very little prerequisite knowledge, so it can be read by students who have not
undertaken a mathematics course for some time. The style is informal, and the text contains
a large number of worked examples. Students are encouraged to tackle problems for themselves
as they read through each section. Detailed solutions are provided so that all answers can be
checked. Consequently, it should be possible to work through this text on a self-study basis.
The material is wide ranging and varies from elementary topics such as percentages and
­linear equations to more sophisticated topics such as constrained optimisation of multivariate
functions. The text should therefore be suitable for use on both low- and high-level quantitative methods courses.
This text was first published in 1991. The prime motivation for writing it then was to
try to produce a text that students could actually read and understand for themselves. This
remains the guiding principle when writing this ninth edition.
One of the main improvements is the inclusion of over 200 additional questions. Each ­chapter
now ends with both multiple choice questions and a selection of longer examination-style
­questions. Students usually enjoy tackling multiple choice questions since they provide a quick
way of testing recall of the material covered in each chapter. Several universities include multiple
choice as part of their assessment. The final section in each chapter entitled “Examination
Questions” contains longer problems which require knowledge and understanding of more than
one topic. Although these have been conveniently placed at the end of each chapter it may be
best to leave these until the end of the academic year so that they can be used during the revision
period just before the examinations.
Ian Jacques
This page intentionally left blank
This page intentionally left blank
INTRODUCTION
Getting Started
NOTES FOR STUDENTS: HOW TO USE THIS TEXT
I am always amazed by the mix of students on first-year economics courses. Some have not
acquired any mathematical knowledge beyond elementary algebra (and even that can be of a
rather dubious nature), some have never studied economics before in their lives, while others
have passed preliminary courses in both. Whatever category you are in, I hope that you will
find this text of value. The chapters covering algebraic manipulation, simple calculus, finance,
matrices and linear programming should also benefit students on business studies and manage­
ment courses.
The first few chapters are aimed at complete beginners and students who have not taken
mathematics courses for some time. I would like to think that these students once enjoyed
mathematics and had every intention of continuing their studies in this area, but somehow never
found the time to fit it into an already overcrowded academic timetable. However, I suspect that
the reality is rather different. Possibly they hated the subject, could not understand it and dropped
it at the earliest opportunity. If you find yourself in this position, you are probably horrified to
discover that you must embark on a quantitative methods course with an examination looming
on the horizon. However, there is no need to worry. My experience is that every student is
capable of passing a mathematics examination. All that is required is a commitment to study
and a willingness to suspend any prejudices about the subject gained at school. The fact that
you have bothered to buy this text at all suggests that you are prepared to do both.
To help you get the most out of this text, let me compare the working practices of
economics and engineering students. The former rarely read individual books in any great depth.
They tend to skim through a selection of books in the university library and perform a large
number of Internet searches, picking out relevant information. Indeed, the ability to read selectively and to compare various sources of information is an important skill that all arts and
social science students must acquire. Engineering students, on the other hand, are more likely
to read just a few books in any one year. They read each of these from cover to cover and
attempt virtually every problem en route. Even though you are most definitely not an engineer,
it is the engineering approach that you need to adopt while studying mathematics. There are
several reasons for this. First, a mathematics text can never be described, even by its most
ardent admirers, as a good bedtime read. It can take an hour or two of concentrated effort to
understand just a few pages of a mathematics text. You are therefore recommended to work
through this text systematically in short bursts rather than to attempt to read whole chapters.
Each section is designed to take between one and two hours to complete, and this is quite
sufficient for a single session. Secondly, mathematics is a hierarchical subject in which one topic
follows on from the next. A construction firm building an office block is hardly likely to erect
the fiftieth storey without making sure that the intermediate floors and foundations are
securely in place. Likewise, you cannot ‘dip’ into the middle of a mathematics text
and expect to follow it unless you have satisfied the prerequisites for that topic. Finally, you
actually need to do mathematics yourself before you can understand it. No matter how
wonderful your lecturer is, and no matter how many problems are discussed in class, it is only
2 INTRODUCTION Getting Started
by solving problems yourself that you are ever going to become confident in using and applying
mathematical techniques. For this reason, several problems are interspersed within the text, and
you are encouraged to tackle these as you go along. You will require writing paper, graph paper,
pens and a calculator for this. There is no need to buy an expensive calculator unless you are
feeling particularly wealthy at the moment. A bottom-of-the-range scientific calculator should
be good enough. Answers to every question are printed at the back of this text so that you
can check your own answers quickly as you go along. However, please avoid the temptation
to look at them until you have made an honest attempt at each one. Remember that in the
future you may well have to sit down in an uncomfortable chair, in front of a blank sheet of
paper, and be expected to produce solutions to examination questions of a similar type.
At the end of each section there are two parallel exercises. The non-starred exercises
are intended for students who are meeting these topics for the first time and the questions are
designed to consolidate basic principles. The starred exercises are more challenging but still
cover the full range so that students with greater experience will be able to concentrate their
efforts on these questions without having to pick-and-mix from both exercises. The chapter
dependence is shown in Figure I.1. If you have studied some advanced mathematics before,
you will discover that parts of Chapters 1, 2 and 4 are familiar. However, you may find that
the sections on economics applications contain new material. You are best advised to test
yourself by attempting a selection of problems from the starred exercise in each section to
see if you need to read through it as part of a refresher course. Economics students in a
desperate hurry to experience the delights of calculus can miss out Chapter 3 without any
loss of continuity and move straight on to Chapter 4. The mathematics of finance is probably
more relevant to business and account­ancy students, although you can always read it later if
it is part of your economics syllabus.
At the end of every chapter you will find a multiple choice test and some examination
questions. These cover the work of the whole chapter. We recommend that you try the
multiple choice questions when you have completed the relevant chapter. As usual, answers
1
Linear equations
7
Matrices
8
Linear programming
2
Non-linear equations
3
Mathematics of finance
4
Differentiation
5
Partial differentiation
6
Integration
9
Dynamics
Figure I.1
INTRODUCTION Getting Started 3
are provided at the back of the book so that you can check to see how well you have done.
If you do get any of the questions wrong, it would be worth re-doing that question perhaps
writing down full working so that you can spot your mistake more easily. The final section
contains several examination-style problems which are more challenging. They tend to be
longer than the questions encountered so far in the exercises and require more confidence
and experience. We recommend that you leave these until the end of the course and use
them in your build-up to the final exams.
I hope that this text helps you to succeed in your mathematics course. You never know,
you might even enjoy it. Remember to wear your engineer’s hat while reading the text. I have
done my best to make the material as accessible as possible. The rest is up to you!
Chapter 1
Linear Equations
The main aim of this chapter is to introduce the mathematics of linear equations. This is an
obvious first choice in an introductory text, since it is an easy topic which has many applications.
There are seven sections, which are intended to be read in the order that they appear.
Sections 1.1, 1.2, 1.3, 1.4 and 1.6 are devoted to mathematical methods. They serve to revise the
rules of arithmetic and algebra, which you probably met at school but may have forgotten. In
particular, the properties of negative numbers and fractions are considered. A reminder is given
on how to multiply out brackets and how to manipulate mathematical expressions. You are also
shown how to solve simultaneous linear equations. Systems of two equations in two unknowns
can be solved using graphs, which are described in Section 1.3. However, the preferred method
uses elimination, which is considered in Section 1.4. This algebraic approach has the advantage
that it always gives an exact solution and it extends readily to larger systems of equations.
The remaining two sections are reserved for applications in microeconomics and macroeconomics.
You may be pleasantly surprised by how much economic theory you can analyse using just the
basic mathematical tools considered here. Section 1.5 introduces the fundamental concept of an
economic function and describes how to calculate equilibrium prices and quantities in supply and
demand theory. Section 1.7 deals with national income determination in simple macroeconomic
models.
The first six sections underpin the rest of the text and are essential reading. The final section is
not quite as important and can be omitted at this stage.
Section 1.1
Introduction to algebra
Objectives
At the end of this section you should be able to:
●
Add, subtract, multiply and divide negative numbers.
●
Understand what is meant by an algebraic expression.
●
Evaluate algebraic expressions numerically.
●
Simplify algebraic expressions by collecting like terms.
●
Multiply out brackets.
●
Factorise algebraic expressions.
Algebra is boring
There is no getting away from the fact that algebra is boring. Doubtless there are a few
enthusiasts who get a kick out of algebraic manipulation, but economics and business students
are rarely to be found in this category. Indeed, the mere mention of the word ‘algebra’ is
enough to strike fear into the heart of many a first-year student. Unfortunately, you cannot
get very far with mathematics unless you have completely mastered this topic. An apposite
analogy is the game of chess. Before you can begin to play a game of chess, it is necessary to
go through the tedium of learning the moves of individual pieces. In the same way it is essential that you learn the rules of algebra before you can enjoy the ‘game’ of mathematics. Of
course, just because you know the rules does not mean that you are going to excel at the
game, and no one is expecting you to become a grandmaster of mathematics. However, you
should at least be able to follow the mathematics presented in economics books and journals
as well as to solve simple problems for yourself.
Advice
If you have studied mathematics recently, then you will find the material in the first few sections of
the text fairly straightforward. You may prefer just to try the questions in the starred exercise at the end
of each section to get yourself back up to speed. However, if it has been some time since you have
studied this subject, our advice is very different. Please work through the material thoroughly even if
it is vaguely familiar. Make sure that you do the problems as they arise, checking your answers with
those provided at the back of this text. The material has been broken down into three subsections:
●
●
●
negative numbers;
expressions;
brackets.
You might like to work through these subsections on separate occasions to enable the ideas to sink
in. To rush this topic now is likely to give you only a half-baked understanding, which will result in
hours of frustration when you study the later chapters of this text.
Section 1.1 Introduction to algebra 7
1.1.1 Negative numbers
In mathematics numbers are classified into one of three types: positive, negative or zero.
At school you were probably introduced to the idea of a negative number via the temperature
on a thermometer scale measured in degrees centigrade. A number such as −5 would then be
interpreted as a temperature of 5 degrees below freezing. In personal finance a negative bank
balance would indicate that an account is ‘in the red’ or ‘in debit’. Similarly, a firm’s profit
of −500 000 signifies a loss of half a million.
The rules for the multiplication of negative numbers are
negative 3 negative 5 positive
negative 3 positive 5 negative
It does not matter in which order two numbers are multiplied, so
positive 3 negative 5 negative
These rules produce, respectively,
(−2) × (−3) = 6
(−4) × 5 = −20
7 × (−5) = −35
Also, because division is the same sort of operation as multiplication (it just undoes the result
of multiplication and takes you back to where you started), exactly the same rules apply when
one number is divided by another. For example,
(−15) ÷ (−3) = 5
(−16) ÷ 2 = −8
2 ÷ (−4) = −1/ 2
In general, to multiply or divide lots of numbers it is probably simplest to ignore the signs
to begin with and just to work the answer out. The final result is negative if the total number
of minus signs is odd and positive if the total number is even.
Example
Evaluate
(a) (−2) × (−4) × (−1) × 2 × (−1) × (−3) (b)
5 3 (–4) 3 (–1) 3 (–3)
(–6) 3 2
Solution
(a) Ignoring the signs gives
2 × 4 × 1 × 2 × 1 × 3 = 48
There are an odd number of minus signs (in fact, five), so the answer is −48.
(b) Ignoring the signs gives
60
5343133
5
55
12
632
There are an even number of minus signs (in fact, four), so the answer is 5.
8 Chapter 1 Linear Equations
Advice
Attempt the following problem yourself both with and without a calculator. On most machines a
negative number such as −6 is entered by pressing the button labelled ( ]) followed by 6.
Practice Problem
1. (1) Without using a calculator, evaluate
(a) 5 × (−6)
(b) (−1) × (−2)
(c) (−50) ÷ 10
(d) (−5) ÷ (−1)
(e) 2 × (−1) × (−3) × 6
(f)
2 3 (–1) 3 (–3) 3 6
(–2) 3 3 3 6
(2) Confirm your answer to part (1) using a calculator.
To add or subtract negative numbers it helps to think in terms of a number line:
]4
]3
]2
]1
0
1
2
3
4
If b is a positive number, then
a−b
can be thought of as an instruction to start at a and to move b units to the left. For example,
1 − 3 = −2
because if you start at 1 and move 3 units to the left, you end up at −2:
]4
]3
]2
]1
0
1
2
3
4
]1
0
1
2
3
4
Similarly,
−2 − 1 = −3
because 1 unit to the left of −2 is −3.
]4
]3
]2
On the other hand,
a − (−b)
is taken to be a + b. This follows from the rule for multiplying two negative numbers, since
−(−b) = (−1) × (−b) = b
Consequently, to evaluate
a − (−b)
Section 1.1 Introduction to algebra 9
you start at a and move b units to the right (that is, in the positive direction). For example,
−2 − (−5) = −2 + 5 = 3
because if you start at −2 and move 5 units to the right, you end up at 3.
]4
]3
]2
0
]1
1
2
3
4
Practice Problem
2. (1) Without using a calculator, evaluate
(a) 1 − 2
(b) −3 − 4
(c) 1 − (−4)
(d) −1 − (−1)
(e) −72 − 19
(f ) −53 − (−48)
(2) Confirm your answer to part (1) using a calculator.
1.1.2 Expressions
In algebra, letters are used to represent numbers. In pure mathematics the most common
letters used are x and y. However, in applications it is helpful to choose letters that are more
meaningful, so we might use Q for quantity and I for investment. An algebraic expression
is then simply a combination of these letters, brackets and other mathematical symbols such
as + or −. For example, the expression
P 1+
r
100
n
can be used to work out how money in a savings account grows over a period of time.
The letters P, r and n represent the original sum invested (called the principal – hence the use
of the letter P), the rate of interest and the number of years, respectively. To work it all out,
you not only need to replace these letters by actual numbers, but you also need to understand
the various conventions that go with algebraic expressions such as this.
In algebra, when we multiply two numbers represented by letters, we usually suppress the
multiplication sign between them. The product of a and b would simply be written as ab
without bothering to put the multiplication sign between the symbols. Likewise, when a
number represented by the letter Y is doubled, we write 2Y. In this case we not only suppress
the multi­plication sign but adopt the convention of writing the number in front of the letter.
Here are some further examples:
P × Q is written as PQ
d × 8 is written as 8d
n × 6 × t is written as 6nt
z × z is written as z2
(using the index 2 to indicate squaring a number)
1 × t is written as t
(since multiplying by 1 does not change a number)
10 Chapter 1 Linear Equations
In order to evaluate these expressions it is necessary to be given the numerical value of
each letter. Once this has been done, you can work out the final value by performing the
operations in the following order:
Brackets first
(B)
Indices second
(I)
Division and Multiplication third
(DM)
Addition and Subtraction fourth
(AS)
This is sometimes remembered using the acronym BIDMAS, and it is essential to use this
ordering for working out all mathematical calculations. For example, suppose you wish to
evaluate each of the following expressions when n = 3:
2n2 and (2n)2
Substituting n = 3 into the first expression gives
2n2 = 2 × 32 (the multiplication sign is revealed when we switch from algebra to numbers)
=2×9
(according to BIDMAS, indices are worked out before multiplication)
= 18
whereas in the second expression we get
(2n)2 = (2 × 3)2
=6
(again, the multiplication sign is revealed)
(according to BIDMAS, we evaluate the inside of the brackets first)
2
= 36
The two answers are not the same, so the order indicated by BIDMAS really does matter.
Looking at the previous list, notice that there is a tie between multiplication and division
for third place, and another tie between addition and subtraction for fourth place. These
pairs of operations have equal priority, and under these circumstances you work from left to
right when evaluating expressions. For example, substituting x = 5 and y = 4 in the expression, x − y + 2, gives
x−y+2=5−4+2
=1+2
(reading from left to right, subtraction comes first)
=3
Example
(a) Find the value of 2x − 3y when x = 9 and y = 4.
(b) Find the value of 2Q2 + 4Q + 150 when Q = 10.
(c) Find the value of 5a − 2b + c when a = 4, b = 6 and c = 1.
(d) Find the value of (12 − t) − (t − 1) when t = 4.
Solution
(a) 2x − 3y = 2 × 9 − 3 × 4
= 18 − 12
=6
(substituting numbers)
(multiplication has priority over subtraction)
Section 1.1 Introduction to algebra 11
(b) 2Q2 + 4Q + 150 = 2 × 102 + 4 × 10 + 150
(substituting numbers)
= 2 × 100 + 4 × 10 + 150(indices have priority over multiplication
and addition)
(multiplication has priority over addition)
= 200 + 40 + 150
= 390
(c) 5a − 2b + c = 5 × 4 − 2 × 6 + 1
(substituting numbers)
= 20 − 12 + 1
(multiplication has priority over addition and
subtraction)
= 8 + 1(addition and subtraction have equal priority, so
work from left to right)
=9
(d) (12 − t) − (t − 1) = (12 − 4) − (4 − 1)
=8−3
(substituting numbers)
(brackets first)
=5
Practice Problem
3. Evaluate each of the following by replacing the letters by the given numbers:
(a) 2Q + 5 when Q = 7.
(b) 5x2y when x = 10 and y = 3.
(c) 4d − 3f + 2g when d = 7, f = 2 and g = 5.
(d) a(b + 2c) when a = 5, b = 1 and c = 3.
Like terms are multiples of the same letter (or letters). For example, 2P, −34P and 0.3P are
all multiples of P and so are like terms. In the same way, xy, 4xy and 69xy are all multiples
of xy and so are like terms. If an algebraic expression contains like terms which are added or
subtracted together, then it can be simplified to produce an equivalent shorter expression.
Example
Simplify each of the following expressions (where possible):
(a) 2a + 5a − 3a
(b) 4P − 2Q
(c) 3w + 9w2 + 2w
(d) 3xy + 2y2 + 9x + 4xy − 8x
Solution
(a) All three are like terms since they are all multiples of a, so the expression can be simplified:
2a + 5a − 3a = 4a
➜
12 Chapter 1 Linear Equations
(b) The terms 4P and 2Q are unlike because one is a multiple of P and the other is a multiple
of Q, so the expression cannot be simplified.
(c) The first and last are like terms since they are both multiples of w, so we can collect
these together and write
3w + 9w2 + 2w = 5w + 9w2
This cannot be simpified any further because 5w and 9w2 are unlike terms.
(d) The terms 3xy and 4xy are like terms, and 9x and 8x are also like terms. These pairs
can therefore be collected together to give
3xy + 2y2 + 9x + 4xy − 8x = 7xy + 2y2 + x
Notice that we write just x instead of 1x and also that no further simplication is possible
since the final answer involves three unlike terms.
Practice Problem
4. Simplify each of the following expressions, where possible:
(a) 2x + 6y − x + 3y
(b) 5x + 2y − 5x + 4z
(c) 4Y2 + 3Y − 43
(d) 8r2 + 4s − 6rs − 3s − 3s2 + 7rs
(e) 2e2 + 5f − 2e2 − 9f
(f) 3w + 6W
(g) ab − ba
1.1.3 Brackets
It is useful to be able to take an expression containing brackets and rewrite it as an equivalent
expression without brackets, and vice versa. The process of removing brackets is called ‘expanding brackets’ or ‘multiplying out brackets’. This is based on the distributive law, which states
that for any three numbers a, b and c
a(b 1 c) 5 ab 1 ac
It is easy to verify this law in simple cases. For example, if a = 2, b = 3 and c = 4, then the
left-hand side is
2(3 + 4) = 2 × 7 = 14
However,
ab = 2 × 3 = 6 and
ac = 2 × 4 = 8
and so the right-hand side is 6 + 8, which is also 14.
This law can be used when there are any number of terms inside the brackets. We have
a(b + c + d) = ab + ac + ad
a(b + c + d + e) = ab + ac + ad + ae
and so on.
Section 1.1 Introduction to algebra 13
It does not matter in which order two numbers are multiplied, so we also have
(b + c)a = ba + ca
(b + c + d)a = ba + ca + da
(b + c + d + e)a = ba + ca + da + ea
Example
Multiply out the brackets in
(a) x(x − 2)
(b) 2(x + y − z) + 3(z + y)
(c) x + 3y − (2y + x)
Solution
(a) The use of the distributive law to multiply out x(x − 2) is straightforward. The x outside
the bracket multiplies the x inside to give x2. The x outside the bracket also multiplies the
−2 inside to give −2x. Hence
x(x − 2) = x2 − 2x
(b) To expand
2(x + y − z) + 3(z + y)
we need to apply the distributive law twice. We have
2(x + y − z) = 2x + 2y − 2z
3(z + y) = 3z + 3y
Adding together gives
2(x + y − z) + 3(z + y) = 2x + 2y − 2z + 3z + 3y
= 2x + 5y + z (collecting like terms)
(c) It may not be immediately apparent how to expand
x + 3y − (2y + x)
However, note that
−(2y + x)
is the same as
(−1)(2y + x)
which expands to give
(−1)(2y) + (−1)x = −2y − x
Hence
x + 3y − (2y + x) = x + 3y − 2y − x = y
after collecting like terms.
14 Chapter 1 Linear Equations
Advice
In this example the solutions are written out in painstaking detail. This is done to show you precisely
how the distributive law is applied. The solutions to all three parts could have been written down in
only one or two steps of working. You are, of course, at liberty to compress the working in your own
solutions, but please do not be tempted to overdo this. You might want to check your answers at a
later date and may find it difficult if you have tried to be too clever.
Practice Problem
5. Multiply out the brackets, simplifying your answer as far as possible.
(a) (5 − 2z)z (b) 6(x − y) + 3(y − 2x) (c) x − y + z − (x2 + x − y)
Mathematical formulae provide a precise way of representing calculations that need to be
worked out in many business models. However, it is important to realise that these formulae
may be valid only for a restricted range of values. Most large companies have a policy to
reimburse employees for use of their cars for travel: for the first 50 miles they may be able
to claim 90 cents a mile, but this could fall to 60 cents a mile thereafter. If the distance,
x miles, is no more than 50 miles, then travel expenses, E (in dollars), could be worked out
using the formula E = 0.9x. If x exceeds 50 miles, the employee can claim $0.90 a mile for the
first 50 miles but only $0.60 a mile for the last (x − 50) miles. The total amount is then
E = 0.9 × 50 + 0.6(x − 50)
= 45 + 0.6x − 30
= 15 + 0.6x
Travel expenses can therefore be worked out using two separate formulae:
●
E = 0.9x
when x is no more than 50 miles
●
E = 15 + 0.6x
when x exceeds 50 miles.
Before we leave this topic, a word of warning is in order. Be careful when removing brackets
from very simple expressions such as those considered in part (c) in the previous worked
example and practice problem. A common mistake is to write
(a + b) − (c + d) = a + b − c + d This is NOT true
The distributive law tells us that the −1 multiplying the second bracket applies to the d as
well as the c, so the correct answer has to be
(a + b) − (c + d) = a + b − c − d
In algebra, it is sometimes useful to reverse the procedure and put the brackets back in.
This is called factorisation. Consider the expression 12a + 8b. There are many numbers which
divide into both 8 and 12. However, we always choose the biggest number, which is 4 in this
case, so we attempt to take the factor of 4 outside the brackets:
12a + 8b = 4(? + ?)
Section 1.1 Introduction to algebra 15
where each ? indicates a mystery term inside the brackets. We would like 4 multiplied by the
first term in the brackets to be 12a, so we are missing 3a. Likewise, if we are to generate an
8b, the second term in the brackets will have to be 2b.
Hence
12a + 8b = 4(3a + 2b)
As a check, notice that when you expand the brackets on the right-hand side, you really do
get the expression on the left-hand side.
Example
Factorise
(a) 6L − 3L2
(b) 5a − 10b + 20c
Solution
(a) Both terms have a common factor of 3. Also, because L2 = L × L, both 6L and −3L2
have a factor of L. Hence we can take out a common factor of 3L altogether.
6L − 3L2 = 3L(2) − 3L(L) = 3L(2 − L)
(b) All three terms have a common factor of 5, so we write
5a − 10b + 20c = 5(a) − 5(2b) + 5(4c) = 5(a − 2b + 4c)
Practice Problem
6. Factorise
(a) 7d + 21 (b) 16w − 20q (c) 6x − 3y + 9z
(d) 5Q − 10Q2
We conclude our discussion of brackets by describing how to multiply two brackets
together. In the expression (a + b)(c + d) the two terms a and b must each multiply the single
bracket (c + d), so
(a + b)(c + d) = a(c + d) + b(c + d)
The first term a(c + d) can itself be expanded as ac + ad. Likewise, b(c + d) = bc + bd. Hence
(a + b)(c + d) = ac + ad + bc + bd
This procedure then extends to brackets with more than two terms:
(a + b)(c + d + e) = a(c + d + e) + b(c + d + e) = ac + ad + ae + bc + bd + be
16 Chapter 1 Linear Equations
Example
Multiply out the brackets
(a) (x + 1)(x + 2) (b) (x + 5)(x − 5) (c) (2x − y)(x + y − 6)
simplifying your answer as far as possible.
Solution
(a) (x + 1)(x + 2) = x(x + 2) + (1)(x + 2)
= x2 + 2x + x + 2
= x2 + 3x + 2
(b) (x + 5)(x − 5) = x(x − 5) + 5(x − 5)
= x2 − 5x + 5x − 25
= x2 − 25
the xs cancel
(c) (2x − y)(x + y − 6) = 2x(x + y − 6) − y(x + y − 6)
= 2x2 + 2xy − 12x − yx − y2 + 6y
= 2x2 + xy − 12x − y2 + 6y
Practice Problem
7. Multiply out the brackets.
(a) (x + 3)(x − 2)
(b) (x + y)(x − y)
(c) (x + y)(x + y)
(d) (5x + 2y)(x − y + 1)
Looking back at part (b) of the previous worked example, notice that
(x + 5)(x − 5) = x2 − 25 = x2 − 52
Quite generally
(a + b)(a − b) = a(a − b) + b(a − b)
= a2 − ab + ba − b2
= a2 − b2
The result
a2 ] b2 5 (a 1 b)(a ] b)
is called the difference of two squares formula. It provides a quick way of factorising certain
expressions.
Section 1.1 Introduction to algebra 17
Example
Factorise the following expressions:
(a) x2 − 16 (b) 9x2 − 100
Solution
(a) Noting that
x2 − 16 = x2 − 42
we can use the difference of two squares formula to deduce that
x2 − 16 = (x + 4)(x − 4)
(b) Noting that
9x2 − 100 = (3x)2 − (10)2 (3x)2 5 3x 3 3x
5 9x 2
we can use the difference of two squares formula to deduce that
9x2 − 100 = (3x + 10)(3x − 10)
Practice Problem
8. Factorise the following expressions:
(a) x2 − 64 (b) 4x2 − 81
Advice
This completes your first piece of mathematics. We hope that you have not found it quite as bad as you
first thought. There now follow a few extra problems to give you more practice. Not only will they help
to strengthen your mathematical skills, but also they should improve your overall confidence. Two
alternative exercises are available. Exercise 1.1 is suitable for students whose mathematics may be
rusty and who need to consolidate their understanding. Exercise 1.1* contains more challenging
problems and so is more suitable for those students who have found this section very easy.
Key Terms
Difference of two squares
The algebraic result which states that a2 − b2 = (a + b)(a − b).
Distributive law The law of arithmetic which states that a(b + c) = ab + ac for any numbers, a, b, c.
Factorisation
brackets.
Like terms
The process of writing an expression as a product of simpler expressions using
Multiples of the same combination of algebraic symbols.
18 Chapter 1 Linear Equations
Exercise 1.1
1. Without using a calculator, evaluate
(a) 10 × (−2)
(e) 24 ÷ (−2)
(b) (−1) × (−3)
(c) (−8) ÷ 2
(d) (−5) ÷ (−5)
(f) (−10) × (−5)
20
(g)
–4
(h)
(i) (−6) × 5 × (−1) ( j)
–27
–9
2 3 (–6) 3 3
(–9)
2. Without using a calculator, evaluate
(a) 5 − 6
(b) −1 − 2
(c) 6 − 17
(d) −7 + 23
(e) −7 − (−6)
(f) −4 − 9
(g) 7 − (−4)
(h) −9 − (−9)
(i) 12 − 43
( j) 2 + 6 − 10
3. Without using a calculator, evaluate
–30 – 6
–18
(a) 5 × 2 − 13
(b)
(e) 1 − 6 × 7
(f) −5 + 6 ÷ 3
(i) (−2)2 − 5 × 6 + 1
(c)
(–3) 3 (–6) 3 (–1)
(d) 5 × (1 − 4)
2–3
(g) 2 × (−3)2
( j)
(h) −10 + 22
(–4)2 3 (–3) 3 (–1)
(–2)3
4. Simplify each of the following algebraic expressions:
(a) 2 × P × Q
(b) I × 8
(c) 3 × x × y
(d) 4 × q × w × z
(e) b × b
(f) k × 3 × k
5. Simplify the following algebraic expressions by collecting like terms:
(a) 6w − 3w + 12w + 4w
(b) 6x + 5y − 2x − 12y
(c) 3a − 2b + 6a − c + 4b − c
(d) 2x2 + 4x − x2 − 2x
(e) 2cd + 4c − 5dc
(f) 5st + s2 − 3ts + t2 + 9
6. Without using a calculator, find the value of the following:
(a) 2x − y when x = 7 and y = 4.
(b) x2 − 5x + 12 when x = 6.
(c) 2m3 when m = 10.
(d) 5fg2 + 2g when f = 2 and g = 3.
(e) 2v + 4w − (4v − 7w) when v = 20 and w = 10.
7. If x = 2 and y = −3, evaluate
(a) 2x + y
(b) x − y
(c) 3x + 4y
(d) xy
(e) 5xy
(f) 4x − 6xy
Section 1.1 Introduction to algebra 19
8. (a) Without using a calculator, work out the value of (−4)2.
(b) Press the following key sequence on your calculator:
(])
4
x2
Explain carefully why this does not give the same result as part (a) and give an
alternative key sequence that does give the correct answer.
9. Without using a calculator, work out
(a) (5 − 2)2
(b) 52 − 22
Is it true in general that (a − b)2 = a2 − b2?
10. Use your calculator to work out the following. Round your answer, if necessary, to
two decimal places.
(a) 5.31 × 8.47 − 1.012
(b) (8.34 + 2.27)/9.41
(c) 9.53 − 3.21 + 4.02
(d) 2.41 × 0.09 − 1.67 × 0.03
(e) 45.76 − (2.55 + 15.83)
(f) (3.45 − 5.38)2
(g) 4.56(9.02 + 4.73)
(h) 6.85/(2.59 + 0.28)
11. Multiply out the brackets:
(a) 7(x − y)
(b) 3(5x − 2y)
(c) 4(x + 3)
(d) 7(3x − 1)
(e) 3(x + y + z)
(f) x(3x − 4)
(g) y + 2z − 2(x + 3y − z)
(a) 25c + 30
(b) 9x − 18
(c) x 2 + 2x
(d) 16x − 12y
(e) 4x2 − 6xy
(f) 10d − 15e + 50
12. Factorise
13. Multiply out the brackets:
(a) (x + 2)(x + 5)
(b) (a + 4)(a − 1)
(c) (d + 3)(d − 8)
(d) (2s + 3)(3s + 7)
(e) (2y + 3)(y + 1)
(f) (5t + 2)(2t − 7) (g) (3n + 2)(3n − 2) (h) (a − b)(a − b)
14. Simplify the following expressions by collecting together like terms:
(a) 2x + 3y + 4x − y
(b) 2x2 − 5x + 9x2 + 2x − 3
(c) 5xy + 2x + 9yx
(d) 7xyz + 3yx − 2zyx + yzx − xy
(e) 2(5a + b) − 4b
(f) 5(x − 4y) + 6(2x + 7y)
(g) 5 − 3(p − 2)
(h) x(x − y + 7) + xy + 3x
15. Use the formula for the difference of two squares to factorise
(a) x2 − 4 (b) Q2 − 49 (c) x2 − y2 (d) 9x2 − 100y2
16. Simplify the following algebraic expressions:
(a) 3x − 4x2 − 2 + 5x + 8x2
(b) x(3x + 2) − 3x(x + 5)
➜
20 Chapter 1 Linear Equations
17. A law firm seeks to recruit top-quality experienced lawyers. The total package offered is
the sum of three separate components: a basic salary which is 1.2 times the candidate’s
current salary together with an additional $3000 for each year worked as a qualified
lawyer and an extra $1000 for every year that they are over the age of 21.
Work out a formula that could be used to calculate the total salary, S, offered to
someone who is A years of age, has E years of relevant experience and who currently
earns $N. Hence work out the salary offered to someone who is 30 years old with five
years’ experience and who currently earns $150 000.
18. Write down a formula for each situation:
(a) A plumber has a fixed call-out charge of $80 and has an hourly rate of $60. Work
out the total charge, C, for a job that takes L hours in which the cost of materials
and parts is $K.
(b) An airport currency exchange booth charges a fixed fee of $10 on all transactions
and offers an exchange rate of 1 dollar to 0.8 euros. Work out the total charge, C,
(in $) for buying x euros.
(c) A firm provides 5 hours of in-house training for each of its semi-skilled workers and
10 hours of training for each of its skilled workers. Work out the total number of
hours, H, if the firm employs a semi-skilled and b skilled workers.
(d) A car hire company charges $C a day together with an additional $c per mile. Work
out the total charge, $X, for hiring a car for d days and travelling m miles during
that time.
Exercise 1.1*
1. Without using a calculator, evaluate
(a) (12 − 8) − (6 − 5)
(b) 12 − (8 − 6) − 5
(c) 12 − 8 − 6 − 5
2. Put a pair of brackets in the left-hand side of each of the following to give correct
statements:
(a) 2 − 7 − 9 + 3 = −17
(b) 8 − 2 + 3 − 4 = −1
(c) 7 − 2 − 6 + 10 = 1
3. Without using a calculator, work out the value of each of the following expressions in
the case when a = 3, b = −4 and c = −2:
(a) a(b − c)
(e)
c+b
2a
(b) 3c(a + b)
(f)
2(b2 – c)
(c) a2 + 2b + 3c
(g)
b
a
–
2c 3b
(d) 2abc2
(h) 5a − b3 − 4c2
4. Without using a calculator, evaluate each of the following expressions in the case when
x = −1, y = −2 and z = 3:
(a) x3 + y2 + z
(b)
x2 1 y2 1 z
x2 1 2xy 2 z
5. Multiply out the brackets and simplify
(x − y)(x + y) − (x + 2)(x − y + 3)
(c)
xyz(x 1 z)(z 2 y)
(x 1 y)(x 2 z)
Section 1.1 Introduction to algebra 21
6. Simplify
(a) x − y − ( y − x)
(b) (x − ((y − x) − y))
(c) x + y − (x − y) − (x − (y − x))
(a) (x + 4)(x − 6)
(b) (2x − 5)(3x − 7)
(c) 2x(3x + y − 2)
(d) (3 + g)(4 − 2g + h)
(e) (2x + y)(1 − x − y)
(f) (a + b + c)(a − b − c)
(a) 9x − 12y
(b) x2 − 6x
(c) 10xy + 15x2
(d) 3xy2 − 6x2y + 12xy
(e) x3 − 2x2
(f) 60x4y6 − 15x2y4 + 20xy3
7. Multiply out the brackets:
8. Factorise
9. Use the formula for the difference of two squares to factorise
(a) p2 − 25 (b) 9c2 − 64 (c) 32v2 − 50d 2 (d) 16x4 − y4
10. Evaluate the following without using a calculator:
(a) 50 5632 − 49 4372
(b) 902 − 89.992
(c) 7592 − 5412
(d) 123 456 7892 − 123 456 7882
11. A specialist paint manufacturer receives $12 for each pot sold. The initial set-up cost for
the production run is $800 and the cost of making each tin of paint is $3.
(a) Write down a formula for the total profit, π, if the firm manufactures x pots of paint
and sells y pots.
(b) Use your formula to calculate the profit when x = 1000 and y = 800.
(c) State any restrictions on the variables in the mathematical formula in part (a).
(d) Simplify the formula in the case when the firm sells all that it manufactures.
12. Factorise
(a) 2KL2 + 4KL
(b) L2 − 0.04K2
(c) K2 + 2LK + L2
Section 1.2
Further algebra
Objectives
At the end of this section you should be able to:
●
Simplify fractions by cancelling common factors.
●
Add, subtract, multiply and divide fractions.
●
Solve equations by doing the same thing to both sides.
●Recognise
●
the symbols <, >, ≤ and ≥.
Solve linear inequalities.
This section is broken down into three manageable subsections:
●
fractions;
●
equations;
●
inequalities.
The advice offered in Section 1.1 applies equally well here. Please try to study these topics
on separate occasions and be prepared to work through the practice problems as they arise
in the text.
1.2.1 Fractions
For a numerical fraction such as
7
8
the number 7, on the top, is called the numerator and the number 8, on the bottom, is called
the denominator. In this text we are also interested in the case when the numerator and
denominator involve letters as well as numbers. These are referred to as algebraic fractions.
For example,
1
2
x 22
and
2x 2 2 1
y1z
are both algebraic fractions. The letters x, y and z are used to represent numbers, so the
rules for the manipulation of algebraic fractions are the same as those for ordinary numerical
fractions. It is therefore essential that you are happy manipulating numerical fractions without
a calculator so that you can extend this skill to fractions with letters.
Two fractions are said to be equivalent if they represent the same numerical value. We know
that 3/4 is equivalent to 6/8 since they are both equal to the decimal number 0.75. It is also
intuitively obvious. Imagine breaking a bar of chocolate into four equal pieces and eating three
Section 1.2 Further algebra 23
of them. You eat the same amount of chocolate as someone who breaks the bar into eight equal
pieces and eats six of them. Each piece is only half the size so you need to compensate by eating
twice as many. Formally, we say that when the numerator and denominator are both multiplied
by the same number, the value of the fraction remains unchanged. In this example we have
3 332 6
5
5
4 432 8
This process can be reversed so equivalent fractions are produced when the numerator and
denominator are both divided by the same number. For example,
16 16/8 2
5
5
24 24/8 3
so the fractions 16/24 and 2/3 are equivalent. A fraction is said to be in its simplest form or
reduced to its lowest terms when there are no factors common to both the numerator and
denominator. To express any given fraction in its simplest form, you need to find the highest
common factor of the numerator and denominator and then divide the top and bottom of
the fraction by this.
Example
Reduce each of the following fractions to its lowest terms:
(a)
14
48
2x
3a
x22
(b) (c)
(d)
(e)
21
60
3xy
6a 1 3b
(x 2 2)(x 1 1)
Solution
(a) The largest number which divides into both 14 and 21 is 7, so we choose to divide top
and bottom by 7:
14 14/7 2
5
5
21 21/7 3
An alternative way of writing this (which will be helpful when we tackle algebraic
fractions) is:
14 2 3 7
2
5
5
21 3 3 7
3
(b) The highest common factor of 48 and 60 is 12, so we write:
48 4 3 12 4
5
5
60 5 3 12 5
(c) The factor x is common to both 2x and 3xy, so we need to divide top and bottom by x;
that is, we cancel the xs:
2x
23x
2
5
5
3xy 3 3 x 3 y
3y
(d) Factorising the denominator gives
6a + 3b = 3(2a + b)
➜
24
chApter 1 LInEAr EqUATIonS
which shows that there is a common factor of 3 in the top and bottom which can be
cancelled:
3a
3a
a
5
5
6a 1 3b 3(2a 1 b) 2a 1 b
(e) We see immediately that there is a common factor of (x − 2) in the top and bottom, so
this can be cancelled:
x22
(x 2 2) (x 1 1)
5
1
x 11
Before we leave this topic, a word of warning is in order. Notice that you can only cancel
by dividing by a factor of the numerator or denominator. In part (d) of the above example
you must not get carried away and attempt to cancel the as, and write something daft like:
a
1
5
2a 1 b 2 1 b
This is NOT true
To see that this is totally wrong, let us try substituting numbers, a = 3, b = 4, say, into both
sides. The left-hand side gives
a
3
3
whereas the right-hand side gives
5
5
2a 1 b
23314
10
1
1
1
5
5 , which is not the same value.
21b
214
6
practice problem
1. Reduce each of the following fractions to its lowest terms:
(a)
9
15
(b)
24
30
(c)
x
2xy
(d)
3x
6x 1 9x2
(e)
x(x 1 1)
x(x 2 4)(x 1 1)
The rules for multiplication and division are as follows:
to multiply fractions you multiply their corresponding numerators and denominators
In symbols,
a c a 3 c ac
3 5
5
b d b 3 d bd
to divide by a fraction you turn it upside down and multiply
In symbols,
a c a d
4 5 3
b d b c
5
ad
bc
turn the divisor
upside down
rule of multiplying
fractions
Section 1.2 FUrTHEr ALgEbrA
example
Calculate
(a)
2 5
3
3 4
(b) 2 3
6
13
(c)
6
4
4
7 21
(d)
1
43
2
Solution
(a) The multiplication rule gives
2 5 2 3 5 10
5
3 5
3 4 3 3 4 12
We could leave the answer like this, although it can be simplified by dividing top and
bottom by 2 to get 5/6. It is also valid to ‘cancel’ by 2 at the very beginning: that is,
1
2
5
135 5
3
5
5
3
42
332 6
(b) The whole number 2 is equivalent to the fraction 2/1, so
23
12
6
2
6
236
5 3
5
5
13 1 13 1 3 13 13
(c) To calculate
6
4
4
7 21
the divisor is turned upside down to get 21/4 and then multiplied to get
3
3
6
4
6
21
333 9
4
3
5
5
5
7 21
71
42
132 2
(d) We write 3 as 3/1, so
1
1 3 1 1 1
435 4 5 3 5
2
2 1 2 3 6
practice problem
2. (1) Without using a calculator, evaluate
(a)
1 3
3
2 4
(b) 7 3
1
4
(c)
2 8
4
3 9
(d)
8
4 16
9
(2) Confirm your answer to part (1) using a calculator.
The rules for addition and subtraction are as follows:
to add (or subtract) two fractions you write them as equivalent fractions
with a common denominator and add (or subtract) their numerators
25
26 Chapter 1 Linear Equations
Example
Calculate
(a)
1 2
1 2
7
5
1 (b) 1 (c)
2
5 5
4 3
12 8
Solution
(a) The fractions 1/5 and 2/5 already have the same denominator, so to add them we just add
their numerators to get
1 2 112 3
5
1 5
5 5
5
5
(b) The fractions 1/4 and 2/3 have denominators 4 and 3. One number that is divisible by
both 3 and 4 is 12, so we choose this as the common denominator. Now 4 goes into
12 exactly 3 times, so
1 133
3
5
5 4 4 3 3 12
multiply top
and bottom by 3
and 3 goes into 12 exactly 4 times, so
2 234
8
5
5 3 3 3 4 12
multiply top
and bottom by 4
Hence
1 2
3
8
3 1 8 11
5
5
1 5
1
4 3 12 12
12
12
(c) The fractions 7/12 and 5/8 have denominators 12 and 8. One number that is divisible by
both 12 and 8 is 24, so we choose this as the common denominator. Now 12 goes into
24 exactly twice, so
7
732
14
5
5
12
24
24
and 8 goes into 24 exactly 3 times, so
5 5 3 3 15
5
5
8
24
24
Hence
7
5 14 15 21
2 5
2
5
12 8 24 24 24
It is not essential that the lowest common denominator is used. Any number will
do provided that it is divisible by the two original denominators. If you are stuck,
then you could always multiply the original two denominators together. In part (c) the
denominators multiply to give 96, so this can be used instead. Now
7
7 3 8 56
5
5
12
96
96
Section 1.2 Further algebra 27
and
5
60
5 3 12
5
5
8
96
96
so
7
5 56 60
56 2 60
24
1
2 5
2
5
5
52
12 8 96 96
96
96
24
as before.
Notice how the final answer to part (c) of this example has been written. We have simply
used the fact that when a negative number is divided by a positive number, the answer is
negative. It is standard practice to write negative fractions like this, so we would write 2
in preference to either
3
4
23
3
23
3
or
and, of course,
is written as .
4
4
24
24
Before we leave this topic a word of warning is in order. Notice that you can add or subtract fractions only after you have gone to the trouble of finding a common denominator. In
particular, the following short-cut does not give the correct answer:
a
c
a1c
This is NOT true
1 5
b d
b1d
As usual you can check for yourself that it is complete rubbish by using actual numbers of
your own choosing.
Practice Problem
3. (1) Without using a calculator, evaluate
(a)
3 1
1 2
7
1
2 (b) 1 (c)
2
7 7
3 5
18 4
(2) Confirm your answer to part (1) using a calculator.
Provided that you can manipulate ordinary fractions, there is no reason why you should
not be able to manipulate algebraic fractions just as easily since the rules are the same.
Example
Find expressions for each of the following:
(a)
x
2
x
x26
2
x11
x
1
(b)
(c) 2
(d)
3
4
1 2
2
x 2 1 x(x 1 4)
x21 x21
x 12
x 12
x12 x11
➜
28 Chapter 1 Linear Equations
Solution
(a) To multiply two fractions we multiply their corresponding numerators and denomi­n­
ators, so
x
2
2x
2
3
5
5
x 2 1 x(x 1 4)
(x 2 1)x(x 1 4) (x 2 1)(x 1 4)
the xs cancel
top and bottom
(b) To divide by
x
x21
we turn it upside down and multiply, so
2
x
2
x21 2
4
5
3
5 x
x
x21 x21
x21
the (x ] 1)s cancel
top and bottom
(c) The fractions
x11
x2 1 2
and
x26
x2 1 2
already have the same denominator, so to add them we just add their numerators
to get
x26
x11
x111x26
2x 2 5
1 2
5
5 2
2
2
x 12
x 12
x 12
x 12
(d) The fractions
x
x12
and
1
x11
have denominators x + 2 and x + 1. An obvious common denominator is given by their
product, (x + 2)(x + 1). Now x + 2 goes into (x + 2)(x + 1) exactly x + 1 times, so
x
x(x 1 1)
5
x 1 2 (x 1 2)(x 1 1)
multiply top and
bottom by (x 1 1)
Also, x + 1 goes into (x + 2)(x + 1) exactly x + 2 times, so
1
(x 1 2)
5
x11
(x 1 2)(x 1 1)
multiply top and
bottom by (x 1 2)
Hence
x
1
x(x 1 1)
(x 1 2)
x(x 1 1) 2 (x 1 2)
2
5
2
5
x 1 2 x 1 1 (x 1 2)(x 1 1) (x 1 2)(x 1 1)
(x 1 2)(x 1 1)
It is worth multiplying out the brackets on the top to simplify: that is,
x2 1 x 2 x 2 2
x2 2 2
5
(x 1 2)(x 1 1)
(x 1 2)(x 1 1)
Section 1.2 FUrTHEr ALgEbrA
29
practice problem
4. Find expressions for the following algebraic fractions, simplifying your answers as far
as possible.
(a)
5
x21
3
x21
x12
(b)
x2
x
4
x 1 10
x11
(c)
4
1
1
x11 x11
(d)
2
1
2
x11 x12
1.2.2 equations
In Section 1.1.2 and again in Section 1.2.1, we have seen how to rewrite an algebraic expression in a simpler but equivalent form. For example, when we write things like
x2 + 3x + 3x2 − 10x = 4x2 − 7x
(collecting like terms)
1
x
x2 2 2
2
5
x 1 2 x 1 1 (x 1 2)(x 1 1)
(part (d) of the previous worked example)
or
we have at the back of our minds the knowledge that the left- and right-hand sides are identical
so that each statement is true for all possible values of x. For this reason the above relations
are called identities. Compare these with statements such as:
7x − 1 = 13
or
x2 − 5x = 1
These relations are called equations and are only true for particular values of x which
need to be found. It turns out that the first equation above has just one solution, whereas the
second has two solutions. The latter is called a quadratic equation and will be considered in
the next chapter.
One naïve approach to the solution of equations such as 7x − 1 = 13 might be to use trial
and error: that is, we could just keep guessing values of x until we find the one that works.
Can you see what x is in this case? However, a more reliable and systematic approach is
to actually solve this equation using the rules of mathematics. In fact, the only rule that we
need is:
you can apply whatever mathematical operation you like to an equation,
provided that you do the same thing to both sides
There is only one exception to this rule: you must never divide both sides by zero. This should
be obvious because a number such as 11/0 does not exist. (If you do not believe this, try
dividing 11 by 0 on your calculator.)
The first obstacle that prevents us from writing down the value of x immediately from the
equation 7x − 1 = 13 is the presence of the −1 on the left-hand side. This can be removed by
adding 1. For this to be legal we must also add 1 to the right-hand side to get
7x − 1 + 1 = 13 + 1
7x = 14
30 Chapter 1 Linear Equations
The second obstacle is the number 7 which is multiplying the x. This can be removed by
dividing the left-hand side by 7. Of course, we must also do the same thing to the right-hand
side to get
7x 14
5
7
7
x=2
This is no doubt the solution that you spotted earlier by simple trial and error, and you may be
wondering why you need to bother with the formal method. The reason is simple: guesswork
will not help to solve more complicated equations in which the solution is non-obvious or
even simple equations in which the solution is a fraction. In these circumstances we need to
follow the approach of ‘balancing the equation’.
Example
Solve
(a) 6x + 1 = 10x − 9
(c)
(b) 3(x − 1) + 2(2x + 1) = 4
20
57
3x 2 1
(d)
9
7
5
x 1 2 2x 1 1
(e)
2x
52
x26
Solution
(a) To solve
6x + 1 = 10x − 9
the strategy is to collect terms involving x on one side of the equation, and to collect
all of the number terms on to the other side. It does not matter which way round this
is done. In this particular case, there are more xs on the right-hand side than there are
on the left-hand side. Consequently, to avoid negative numbers, you may prefer to
stack the x terms on the right-hand side. The details are as follows:
1 5 4x 2 9
10 5 4x
10
5x
4
(subtract 6x from both sides)
(add 9 to both sides)
(divide both sides by 4)
Hence x = 5/2 = 21/2.
(b) The novel feature of the equation
3(x − 1) + 2(2x + 1) = 4
is the presence of brackets. To solve it, we first remove the brackets by multiplying out,
and then collect like terms:
3x − 3 + 4x + 2 = 4 (multiply out the brackets)
7x − 1 = 4 (collect like terms)
Note that this equation is now of the form that we know how to solve:
7x 5 5
5
x5
7
(add 1 to both sides)
(divide both sides by 7)
Section 1.2 Further algebra 31
(c) The novel feature of the equation
20
57
3x 2 1
is the fact that it involves an algebraic fraction. This can easily be removed by multi­
plying both sides by the bottom of the fraction:
20
3 (3x 2 1) 5 7(3x 2 1)
3x 2 1
which cancels down to give
20 = 7(3x − 1)
The remaining steps are similar to those in part (b):
(multiply out the brackets)
(add 7 to both sides)
20 5 21x 2 7
27 5 21x
27
5x
21
(divide both sides by 21)
Hence x = 9/7 = 12/7.
(d) The next equation,
9
7
5
x 1 2 2x 1 1
looks particularly daunting since there are fractions on both sides. However, these are
easily removed by multiplying both sides by the denominators, in turn:
7(x 1 2)
2x 1 1
9(2x + 1) = 7(x + 2)
95
(multiply both sides by x + 2)
(multiply both sides by 2x + 1)
With practice you can do these two steps simultaneously and write this as the first line
of working. The procedure of going straight from
9
7
5
x 1 2 2x 1 1
to
9(2x + 1) = 7(x + 2)
is called ‘cross-multiplication’. In general, if
a
c
5
b d
then
a
b
c
d
ad = bc
The remaining steps are similar to those used in the earlier parts of this example:
18x + 9 = 7x + 14
(multiply out the brackets)
11x + 9 = 14
(subtract 7x from both sides)
➜
32 Chapter 1 Linear Equations
(subtract 9 from both sides)
11x = 5
5
x5
11
(divide both sides by 11)
(e) The left-hand side of the final equation
2x
52
x26
is surrounded by a square root, which can easily be removed by squaring both sides
to get
2x
54
x26
The remaining steps are ‘standard’:
2x = 4(x − 6)
(multiply both sides by x − 6)
2x = 4x − 24
(multiply out the brackets)
−2x = −24
(subtract 4x from both sides)
(divide both sides by −2)
x = 12
Looking back over each part of the previous example, notice that there is a common strategy.
In each case, the aim is to convert the given equation into one of the form
ax + b = c
which is the sort of equation that we can easily solve. If the original equation contains brackets,
then remove them by multiplying out. If the equation involves fractions, then remove them
by cross-multiplying.
Advice
If you have the time, it is always worth checking your answer by substituting your solution back
into the original equation. For the last part of the above example, putting x = 12 into
2x
gives
x2 6
2 3 12
24
5
5 4 5 2 ✓
12 2 6
6
Practice Problem
5. Solve each of the following equations. Leave your answer as a fraction, if necessary.
(a) 4x + 1 = 25
(d)
4
55
x21
(b) 4x + 5 = 5x − 7
(e)
3
5
5
x x21
(c) 3(3 − 2x) + 2(x − 1) = 10
Section 1.2 Further algebra 33
1.2.3 Inequalities
In Section 1.1.1 we made use of a number line:
]5
]4
]3
]2
0
]1
1
2
3
4
5
Although only whole numbers are marked on this diagram, it is implicitly assumed that it can
also be used to indicate fractions and decimal numbers. Each point on the line corresponds to
a particular number. Conversely, every number can be represented by a particular point on
the line. For example, −21/2 lies exactly halfway between −3 and −2. Similarly, 47/8 lies 7/8 of
the way between 4 and 5. In theory, we can even find a point on the line corresponding to a
number such as 2 , although it may be difficult to sketch such a point accurately in practice.
My calculator gives the value of 2 to be 1.414 213 56 to eight decimal places. This number
therefore lies just less than halfway between 1 and 2.
1
]5
]4
]3
7
2
]2 2
]2
]1
0
1
48
2
3
4
5
A number line can be used to decide whether or not one number is greater or less than
another number. We say that a number a is greater than a number b if a lies to the right of b
on the line, and we write this as
a>b
Likewise, we say that a is less than b if a lies to the left of b, and we write this as
a<b
From the diagram we see that
−2 > −4
because −2 lies to the right of −4. This is equivalent to the statement
−4 < −2
Similarly,
0 > −1
(or equivalently −1 < 0)
2 > −2 /2 (or equivalently −21/2 < 2)
1
4 7/ 8 . 2 (or equivalently 2 , 4 7/ 8)
There are occasions when we would like the letters a and b to stand for mathematical
expressions rather than actual numbers. In this situation we sometimes use the symbols ≥
and ≤ to mean ‘greater than or equal to’ and ‘less than or equal to’, respectively.
We have already seen that we can manipulate equations in any way we like, provided that
we do the same thing to both sides. An obvious question to ask is whether this rule extends
to inequalities.
34
chApter 1 LInEAr EqUATIonS
Consider the true statement
1<3
(*)
●
Adding 4 to both sides gives 5 < 7, which is true.
●
Adding −5 to both sides gives −4 < −2, which is true.
●
Multiplying both sides by 2 gives 2 < 6, which is true.
However,
●
Multiplying both sides by −6 gives −6 < −18, which is false. In fact, quite the reverse is
true; −6 is actually greater than −18. This indicates that the rule needs modifying before
we can extend it to inequalities and that we must be careful when manipulating inequalities.
practice problem
6. Starting with the true statement
6>3
decide which of the following are valid operations when performed on both sides:
(a) add 6
(b) multiply by 2
(c) subtract 3
(d) add −3
(e) divide by 3
(f) multiply by −4
(g) multiply by −1
(h) divide by −3
(i) add −10
These examples show that the usual rule does apply to inequalities with the important
proviso that
if both sides are multiplied or divided by a negative
number, then the sense of the inequality is reversed
By this we mean that ‘>’ changes to ‘<’, ‘≤’ changes to ‘≥’ and so on.
To see how this works out in practice, consider the inequality
2x + 3 < 4x + 7
If we try to solve this like we did for equations, the first step would be to subtract 4x from
both sides to get
−2x + 3 < 7
and then take 3 away from both sides to get
−2x < 4
Finally, we divide both sides by −2 to get
x > −2
Notice that the sense has been reversed at this stage because we have divided by a negative
number.
Section 1.2 Further algebra 35
Advice
You should check your answer using a couple of test values. Substituting x = 1 (which lies to the
right of −2 and so should work) into both sides of the original inequality 2x + 3 < 4x + 7 gives
5 < 11, which is true. On the other hand, substituting x = −3 (which lies to the left of −2 and so
should fail) gives −3 < −5, which is false. Of course, just checking a couple of numbers like this does
not prove that the final inequality is correct, but it should protect you against gross blunders.
Practice Problem
7. Simplify the inequalities
(a) 2x < 3x + 7 (b) 21x − 19 ≥ 4x + 15
Inequalities arise in business when there is a budgetary restriction on resource allocation.
The following example shows how to set up and solve the relevant inequality.
Example
A firm’s Human Resources department has a budget of $25 000 to spend on training and
laptops. Training courses cost $700 and new laptops are $1200.
(a) If the department trains E employees and buys L laptops, write down an inequality for
E and L.
(b) If 12 employees attend courses, how many laptops could be bought?
Solution
(a) The cost of training E employees is 700E and the cost of buying L laptops is 1200L. The
total amount spent must not exceed $25 000, so 700E + 1200L ≤ 25 000.
(b) Substituting E = 12 into the inequality gives 8400 + 1200L ≤ 25 000.
1200L ≤ 16 600
L # 13
5
6
(subtract 8400 from both sides)
(divide both sides by 1200)
so a maximum of 13 laptops could be bought.
36 Chapter 1 Linear Equations
Key Terms
Algebraic fraction Ratio of two expressions; p(x)/q(x) where p(x) and q(x) are algebraic expressions such as ax2 + bx + c or dx + e.
Denominator The number (or expression) on the bottom of a fraction.
Equation
Equality of two algebraic expressions which is true only for certain values of the variable.
Fractions which may appear different but which have the same numerical
Equivalent fractions
value.
Factor Part of an expression which, when multiplied by all the other factors, gives the complete
expression.
Identity Equality of two algebraic expressions which is true for all values of the variable.
Number line An infinite line on which the points represent real numbers by their (signed) distance
from the origin.
Numerator The number (or expression) on the top of a fraction.
Exercise 1.2
1. Reduce each of the following numerical fractions to their lowest terms:
(a)
13
26
(b)
9
12
(c)
18
30
(d)
24
72
(e)
36
27
2. In 2011 in the United States, 35 out of every 100 adults owned a smartphone. By 2013
this figure increased to 56 out of every 100.
(a) Express both of these figures as fractions reduced to their lowest terms.
(b) By what factor did smartphone ownership increase during this period? Give your
answer as a mixed fraction in its lowest terms.
3. Reduce each of the following algebraic fractions to their lowest terms:
(a)
6x
9
(b)
x
2x 2
(c)
b
abc
(d)
4x
6x 2y
(e)
15a 2b
20ab2
4. By factorising the numerators and/or denominators of each of the following fractions,
reduce each to its lowest terms:
(a)
2p
4q 1 6r
(b)
x
x 2 4x
2
(c)
3ab
6a 1 3a
2
(d)
14d
21d 2 7de
(e)
x12
x2 2 4
5. Which one of the following algebraic fractions can be simplified? Explain why the other
two fractions cannot be simplified.
x]1
,
2x ] 2
x22
,
x12
5t
10t 2 s
Section 1.2 Further algebra 37
6. (1) Without using a calculator, work out the following, giving your answer in its lowest
terms:
(a)
1 2
1
7 7
(b)
2 5
]
9 9
(c)
1 1
1
2 3
(d)
3 2
]
4 5
(e)
1 2
1
6 9
(f)
1 2
1
6 3
(g)
5 3
3
6 4
(h)
4
2
4
15 3
(i)
7 2
3
8 3
( j)
2
4
4
75 5
(k)
2
43
9
(l) 3 4
2
7
(2) Use your calculator to check your answers to part (1).
7. It takes 11/4 hours to complete an annual service of a car. If a garage has 471/2 hours
available, how many cars can it service?
8. Work out each of the following, simplifying your answer as far as possible:
(a)
2
1
1
3x 3x
(b)
2
x
3
x
5
(c)
3
2
2 2
x
x
(d)
7 2
1
x
y
(e)
a a
4
2 6
(f)
5c 5d
1
12 18
(g)
x12
y25
3
y25
x13
(h)
4gh
2g
4
7
9h
(i)
t
45
4
( j)
P Q
3
Q
P
9. Solve each of the following equations. If necessary give your answer as a mixed fraction
reduced to its lowest terms.
(a) x + 2 = 7
(b) 3x = 18
(e) 2x − 3 = 17
(f) 3x + 4 = 1
(i) 4 − x = 9
(j) 6x + 2 = 5x − 1
(m)
4x 2 7
52
3
(n)
4
51
x11
x
52
9
x
(g) 2 7 5 3
6
(c)
(k) 5(3x + 8) = 10
(o) 5 2
(d) x − 4 = −2
(h) 3(x − 1) = 2
(l) 2(x − 3) = 5(x + 1)
1
51
x
10. Which of the following inequalities are true?
(a) −2 < 1
(b) −6 > −4
(c) 3 < 3
(d) 3 ≤ 3
(e) −21 ≥ −22
(f) 4 < 25
11. Simplify the following inequalities:
(a) 2x > x + 1
(b) 7x + 3 ≤ 9 + 5x
(c) x − 5 > 4x + 4
12. Simplify the following algebraic expression:
4
2x
4
x 2y
y
13. (a) Solve the equation
6(2 + x) = 5(1 − 4x)
(b) Solve the inequality
3x + 6 ≥ 5x − 14
(d) x − 1 < 2x − 3
38 Chapter 1 Linear Equations
Exercise 1.2*
1. Simplify each of the following algebraic fractions:
(a)
2x 2 6
4
(b)
9x
6x2 2 3x
(c)
4x 1 16
x14
(d)
x21
12x
(e)
x16
x 2 2 36
(f)
(x 1 3)(2x 2 5)
(2x 2 5)(x 1 4)
(g)
3x
3
6x 2 15x2 1 9x
(h)
4x2 2 25y2
6x 2 15y
(d)
2
1
8
3
4
5 12 25
2. (1) Without using your calculator, evaluate
(a)
4 25
3
5 28
(b)
2 14
5
3
3
7 25 8
(c)
9
3
4
16 8
(e)
10 12
2
13 13
(f)
5 2
1
9 3
3
3
(g) 2 1 1
7
5
3
3
(i) 3 3 1
5
4
( j)
3
1 1
3 2 1
5
3 2
(k)
(h) 5
5
1
2
3 2 21
6
3
5
(l)
2
9
1
2 11
5
10 2
1
1
5
3 42 4
3
6
13
(2) Confirm your answer to part (1) using a calculator.
3. Find simplified expressions for the following fractions:
2
3
1
xy
xy
(a)
x2 1 6x
x22
3
x22
x
(b)
1
1
4
x x11
(c)
(e)
3
4
1
x x11
(f)
3
5
1 2
x
x
(g) x 2
2
x11
(d)
x
x11
1
2
3
(h)
5
2
3
2 1
x(x 1 1)
x x11
4. Solve the following equations:
(a) 5(2x + 1) = 3(x − 2)
(b) 5(x + 2) + 4(2x − 3) = 11
(c) 5(1 − x) = 4(10 + x)
(d) 3(3 − 2x) − 7(1 − x) = 10
(e) 9 − 5(2x − 1) = 6
(f)
3
52
2x 1 1
(h)
x
1357
2
( j)
5(x 2 3) 2(x 2 1)
5
2
5
(g)
2
3
5
x 2 1 5x 1 4
(i) 5 2
(k)
x
52
3
(2x 2 5) 5 3
(m) (x + 2)2 + (2x − 1)2 = 5x(x + 1)
(o)
45
53
2x 2 1
(l) (x + 3)(x − 1) = (x + 4)(x − 3)
(n)
2x 1 7 x 2 4 1
5
1
2
3
6
(p)
4 3
1
2 5
x 4 4x
5. Two-thirds of Ariadne’s money together with five-sevenths of Brian’s money is equal
to three-fifths of Catriona’s money. If Ariadne has $2.40 and Catriona has $11.25,
write down an equation that you could use to work out how much Brian has. Solve this
equation.
Section 1.2 Further algebra 39
6. An amount $P is placed in a savings account. The interest rate is r% compounded
annually so that after n years the savings, S, will be
S 5 P 11
r
100
n
(a) Find S when P = 2000, n = 5 and r = 10.
(b) Find P when S = 65 563.62, n = 3 and r = 3.
(c) Find r when S = 7320.50, P = 5000 and n = 4.
7. Solve the following inequalities:
(a) 2x − 19 > 7x + 24
(d) 3 1
x
, 2(x 1 4)
3
(b) 2(x − 1) < 5(3x + 2)
(c)
2x 2 1
x23
$
5
2
(e) x < 2x + 1 ≤ 7
8. The design costs of an advertisement in a glossy magazine are $9000 and the cost per cm2
of print is $50.
(a) Write down an expression for the total cost of publishing an advert which covers x cm2.
(b) The advertising budget is between $10 800 and $12 500. Write down and solve an
inequality to work out the minimum and maximum area that could be used.
9. List all the whole numbers that satisfy both of the following inequalities simultaneously:
−7 ≤ 2x < 6 and 4x + 1 ≤ x + 2
10. (a) Simplify
31x 2 8
14
2
(2x 2 1)(x 1 2) x 1 2
(b) Solve the equation
x11 x13 1
2
5
8
4
2
(c) Simplify the inequality
(2x + 1)(x − 5) ≤ 2(x + 2)(x − 4)
11. Simplify
x2
2x
4
x 1 1 x2 2 1
Section 1.3
Graphs of linear equations
Objectives
At the end of this section you should be able to:
●
Plot points on graph paper given their coordinates.
●
Sketch a line by finding the coordinates of two points on the line.
●
Solve simultaneous linear equations graphically.
●
Sketch a line by using its slope and intercept.
Consider the two straight lines shown in Figure 1.1. The horizontal line is referred to as the
x axis and the vertical line is referred to as the y axis. The point where these lines intersect is
known as the origin and is denoted by the letter O. These lines enable us to identify uniquely
any point, P, in terms of its coordinates (x, y). The first number, x, denotes the horizontal
distance along the x axis and the second number, y, denotes the vertical distance along the y
axis. The arrows on the axes indicate the positive direction in each case.
y
P (x, y)
y
x
O
x
Figure 1.1
Figure 1.2 shows the five points A(2, 3), B(−1, 4), C(−3, −1), D(3, −2) and E(5, 0) plotted
on coordinate axes. The point A with coordinates (2, 3) is obtained by starting at the origin,
moving 2 units to the right and then moving 3 units vertically upwards. Similarly, the point B
with coordinates (−1, 4) is located 1 unit to the left of O (because the x coordinate is negative) and 4 units up.
Note that the point C lies in the bottom left-hand quadrant since its x and y coordinates are
both negative. It is also worth noticing that E actually lies on the x axis since its y coordinate
is zero. Likewise, a point with coordinates of the form (0, y) for some number y would lie
somewhere on the y axis. Of course, the point with coordinates (0, 0) is the origin, O.
Section 1.3 Graphs of linear equations 41
y
5
B
4
A
3
2
1
E
]5
]4
]3
]2
]1
C
O
1
2
3
4
5
]1
D
]2
]3
]4
]5
Figure 1.2
Practice Problem
1. Plot the following points on graph paper. What do you observe?
(2, 5), (1, 3), (0, 1), (−2, −3), (−3, −5)
In economics we need to do rather more than just plot individual points on graph paper.
We would like to be able to sketch curves represented by equations and to deduce information
from such a picture. We restrict our attention in this section to those equations whose graphs
are straight lines, deferring consideration of more general curve sketching until Chapter 2.
In Practice Problem 1 you will have noticed that the five points (2, 5), (1, 3), (0, 1),
(−2, −3) and (−3, −5) all lie on a straight line. In fact, the equation of this line is
−2x + y = 1
Any point lies on this line if its x and y coordinates satisfy this equation. For example, (2, 5)
lies on the line because when the values x = 2, y = 5 are substituted into the left-hand side of
the equation, we obtain
−2(2) + 5 = −4 + 5 = 1
which is the right-hand side of the equation. The other points can be checked similarly (see Table 1.1).
Table 1.1
Point
Check
(1, 3)
(0, 1)
(−2, −3)
(−3, −5)
−2(1) + 3 = −2 + 3 = 1
−2(0) + 1 = 0 + 1 = 1
−2(−2) − 3 = 4 − 3 = 1
−2(−3) − 5 = 6 − 5 = 1
✓
✓
✓
✓
42 Chapter 1 Linear Equations
The general equation of a straight line takes the form
a multiple of x 1 a multiple of y 5 a number
that is,
dx + ey = f
for some given numbers d, e and f. Consequently, such an equation is called a linear equation.
The numbers d and e are referred to as the coefficients. The coefficients of the linear equation,
−2x + y = 1
are −2 and 1 (the coefficient of y is 1 because y can be thought of as 1 × y).
Practice Problem
2. Check that the points
(−1, 2), (−4, 4), (5, −2), (2, 0)
all lie on the line
2x + 3y = 4
and hence sketch this line on graph paper. Does the point (3, −1) lie on this line?
In general, to sketch a line from its mathematical equation, it is sufficient to calculate
the coordinates of any two distinct points lying on it. These two points can be plotted on
graph paper and a ruler used to draw the line passing through them. One way of finding the
coordinates of a point on a line is simply to choose a numerical value for x and to substitute
it into the equation. The equation can then be used to deduce the corresponding value of y.
The whole process can be repeated to find the coordinates of the second point by choosing
another value for x.
Example
Sketch the line
4x + 3y = 11
Solution
For the first point, let us choose x = 5. Substitution of this number into the equation gives
4(5) + 3y = 11
20 + 3y = 11
The problem now is to solve this equation for y:
3y = −9
y = −3
(subtract 20 from both sides)
(divide both sides by 3)
Consequently, the coordinates of one point on the line are (5, −3).
Section 1.3 Graphs of linear equations 43
For the second point, let us choose x = −1. Substitution of this number into the equation
gives
4(−1) + 3y = 11
−4 + 3y = 11
This can be solved for y as follows:
3y = 15
y=5
(add 4 to both sides)
(divide both sides by 3)
Hence (−1, 5) lies on the line, which can now be sketched on graph paper as shown in
Figure 1.3.
y
5
(]1, 5)
4
3
4x 1 3y 5 11
2
1
]5
]4
]3
]2
1
]1
2
3
4
5
x
]1
]2
]3
(5, ]3)
]4
]5
Figure 1.3
Practice Problem
3. Find the coordinates of two points on the line
3x − 2y = 4
by taking x = 2 for the first point and x = −2 for the second point. Hence sketch its graph.
In this example we arbitrarily picked two values of x and used the linear equation to work
out the corresponding values of y. There is nothing particularly special about the variable x.
We could equally well have chosen values for y and solved the resulting equations for x. In
fact, the easiest thing to do (in terms of the amount of arithmetic involved) is to put x = 0
and find y and then to put y = 0 and find x.
44 Chapter 1 Linear Equations
Example
Sketch the line
2x + y = 5
Solution
Setting x = 0 gives
2(0) + y = 5
0+y=5
y=5
Hence (0, 5) lies on the line.
Setting y = 0 gives
2x + 0 = 5
2x = 5
x = 5/2 (divide both sides by 2)
Hence (5/2, 0) lies on the line.
The line 2x + y = 5 is sketched in Figure 1.4. Notice how easy the algebra is using this
approach. The two points themselves are also slightly more meaningful. They are the points
where the line intersects the coordinate axes.
y
y intercept
5
(0, 5)
4
2x 1 y 5 5
3
2
1
(2.5, 0)
]5
]4
]3
]2
1
]1
2
]1
]2
x intercept
]3
]4
]5
Figure 1.4
3
4
5
x
Section 1.3 Graphs of linear equations 45
Practice Problem
4. Find the coordinates of the points where the line
x − 2y = 2
intersects the axes. Hence sketch its graph.
In economics it is sometimes necessary to handle more than one equation at the same time.
For example, in supply and demand analysis we are interested in two equations, the supply
equation and the demand equation. Both involve the same variables Q and P, so it makes
sense to sketch them on the same diagram. This enables the market equilibrium quantity and
price to be determined by finding the point of intersection of the two lines. We shall return
to the analysis of supply and demand in Section 1.5. There are many other occasions in
economics and business studies when it is necessary to determine the coordinates of points
of intersection. The following is a straightforward example which illustrates the general
principle.
Example
Find the point of intersection of the two lines
4x + 3y = 11
2x + y = 5
Solution
We have already seen how to sketch these lines in the previous two examples. We discovered
that
4x + 3y = 11
passes through (5, −3) and (−1, 5), and that
2x + y = 5
passes through (0, 5) and (5/2, 0).
These two lines are sketched on the same diagram in Figure 1.5, from which the point
of intersection is seen to be (2, 1).
It is easy to verify that we have not made any mistakes by checking that (2, 1) lies on
both lines.
It lies on 4x + 3y = 11 because 4(2) + 3(1) = 8 + 3 = 11 ✓
and lies on 2x + y = 5 because 2(2) + 1 = 4 + 1 = 5. ✓
For this reason, we say that x = 2, y = 1 is the solution of the simultaneous linear equations
4x + 3y = 11
2x + y = 5
➜
46 Chapter 1 Linear Equations
y
5
4
3
point of
intersection
2
1
]5
]4
]3
]2
(2, 1)
1
]1
2
3
4
x
5
]1
4x 1 3y 5 11
]2
]3
2x 1 y 5 5
]4
]5
Figure 1.5
Practice Problem
5. Find the point of intersection of
3x − 2y = 4
x − 2y = 2
[Hint: you might find your answers to Problems 3 and 4 useful.]
Quite often it is not necessary to produce an accurate plot of an equation. All that may be
required is an indication of the general shape together with a few key points or features. It
can be shown that, provided e is non-zero, any equation given by
dx + ey = f
can be rearranged into the special form
y = ax + b
An example showing you how to perform such a rearrangement will be considered in a
moment. The coefficients a and b have particular significance, which we now examine. To be
specific, consider
y = 2x − 3
in which a = 2 and b = −3.
Section 1.3 Graphs of linear equations 47
When x is taken to be zero, the value of y is
y = 2(0) − 3 = −3
The line passes through (0, −3), so the y intercept is −3. This is just the value of b. In other
words, the constant term, b, represents the intercept on the y axis.
In the same way it is easy to see that a, the coefficient of x, determines the slope of a line.
The slope of a straight line is simply the change in the value of y brought about by a 1 unit
increase in the value of x. For the equation
y = 2x − 3
let us choose x = 5 and increase this by a single unit to get x = 6. The corresponding values
of y are then, respectively,
y = 2(5) − 3 = 10 − 3 = 7
y = 2(6) − 3 = 12 − 3 = 9
The value of y increases by 2 units when x rises by 1 unit. The slope of the line is therefore
2, which is the value of a. The slope of a line is fixed throughout its length, so it is immaterial
which two points are taken. The particular choice of x = 5 and x = 6 was entirely arbitrary.
You might like to convince yourself of this by choosing two other points, such as x = 20 and
x = 21, and repeating the previous calculations.
A graph of the line
y = 2x − 3
is sketched in Figure 1.6. This is sketched using the information that the intercept is −3 and
that for every 1 unit along we go 2 units up. In this example the coefficient of x is positive.
This does not have to be the case. If a is negative, then for every increase in x there is a
y
3
2
2 units
1
]2
1 unit
1
]1
2
]1
y 5 2x ] 3
]2
intercept
Figure 1.6
]3
3
x
48 Chapter 1 Linear Equations
y
positive slope
b
zero slope
x
negative slope
Figure 1.7
corresponding decrease in y, indicating that the line is downhill. If a is zero, then the equation
is just
y=b
indicating that y is fixed at b and the line is horizontal. The three cases are illustrated in
Figure 1.7.
It is important to appreciate that in order to use the slope–intercept approach, it is necessary for the equation to be written as
y = ax + b
If a linear equation does not have this form, it is usually possible to perform a preliminary
rearrangement to isolate the variable y on the left-hand side.
For example, to use the slope–intercept approach to sketch the line
2x + 3y = 12
we begin by removing the x term from the left-hand side. Subtracting 2x from both sides gives
3y = 12 − 2x
and dividing both sides by 3 gives
y 5 4 2 23 x
This is now in the required form with a = −2/3 and b = 4. The line is sketched in Figure 1.8.
A slope of −2/3 means that, for every 1 unit along, we go 2/3 units down (or, equivalently,
for every 3 units along, we go 2 units down). An intercept of 4 means that it passes through
(0, 4).
Practice Problem
6. Use the slope–intercept approach to sketch the lines
(a) y = x + 2
(b) 4x + 2y = 1
Section 1.3 Graphs of linear equations 49
y
5
4
intercept
3 units
3
2 units
2
1
1
]1
2
3
4
]1
5
6
y54]
7
x
2
x
3
Figure 1.8
We conclude this section with two examples showing how linear graphs can be used in business.
Example
Two new models of a smartphone are launched on 1 January 2018. Predictions of sales are
given by:
Model 1: S1 = 4 + 0.5n Model 2: S2 = 8 + 0.1n
where Si (in tens of thousands) denotes the monthly sales of model i after n months.
(a) State the values of the slope and intercept of each line and give an interpretation.
(b) Illustrate the sales of both models during the first year by drawing graphs on the same
axes.
(c) Use the graph to find the month when sales of Model 1 overtake those of Model 2.
Solution
(a) The intercept for Model 1 is 4. There are 40 000 sales of this phone when the product
is launched. The slope is 0.5, so each month, sales increase by 5000. The corresponding
figures for Model 2 are 8 and 0.1, respectively.
(b) The intercept for Model 1 is 4, so the line passes through (0, 4). For every one-unit
increase in n the value of S1 increases by 0.5. So, for example, a two-unit increase in n
results in a one-unit increase in S1. The line passes through (2, 5), (4, 6) and so on. The
line is sketched in Figure 1.9.
For Model 2, the line passes through (0, 8), and since the slope is 0.1, it passes
through (10, 9).
(c) The graphs intersect at (10, 9), so sales of Model 1 overtake sales of Model 2 after
10 months.
➜
50 Chapter 1 Linear Equations
12
11
10
Model 2
(10, 9)
Sales (in 10 000s)
9
8
7
6
Model 1
5
4
3
2
1
1
2
3
4
5
6
7
8
9
10
11
12
Time (months)
Figure 1.9
Example
Three companies can supply a university with some mathematical software. Each company
has a different pricing structure:
Company 1 provides a site licence which costs $130 000 and can be used by anyone at
the university;
Company 2 charges $1000 per user;
Company 3 charges a fixed amount of $40 000 for the first 60 users and $500 for each
additional user.
(a) Draw a graph of each cost function on the same set of axes.
(b) What advice can you give the university about which company to use?
Solution
(a) If there are n users, then the cost, C, from each supplier is:
Company 1: C = 130 000. The graph is a horizontal line with intercept 130 000
Company 2:C = 1000n. The graph is a line passing through the origin with a slope
1000
Company 3: If n ≤ 60 then C = 40 000
If n > 60 then C = 40 000 + 500(n − 60) = 500n + 10 000
The graph for Company 3 is a horizontal line with intercept 40 000 until n = 60 after
which the line bends upwards with a slope 500.
The graphs are sketched in Figure 1.10.
Section 1.3 Graphs of linear equations 51
Company 2
240 000
Cost
Company 3
130 000
Company 1
40 000
40 60
Number of users
Figure 1.10
(b) The cheapest supplier depends on the number of users:
If n ≤ 40, Company 2 is the cheapest.
If 40 ≤ n ≤ 240, Company 3 is the cheapest.
If n ≥ 240, Company 1 is the cheapest.
Key Terms
Coefficient A numerical multiplier of the variables in an algebraic term, such as the numbers 4
and 7 in the expression 4x + 7yz2.
Coordinates
Intercept
A set of numbers that determine the position of a point relative to a set of axes.
The point(s) where a graph crosses one of the coordinate axes.
Linear equation
An equation of the form dx + ey = f.
Origin The point where the coordinate axes intersect.
Simultaneous linear equations A set of linear equations in which there are (usually) the same
number of equations and unknowns. The solution consists of values of the unknowns which satisfy
all of the equations at the same time.
Slope of a line Also known as the gradient, it is the change in the value of y when x increases by
1 unit.
x axis The horizontal coordinate axis pointing from left to right.
y axis The vertical coordinate axis pointing upwards.
52 Chapter 1 Linear Equations
Exercise 1.3
1. On graph paper draw axes with values of x and y between −3 and 10, and plot the
following points:
P(4, 0), Q(−2, 9), R(5, 8), S(−1, −2)
Hence find the coordinates of the point of intersection of the line passing through P
and Q, and the line passing through R and S.
2. An airline charges $300 for a flight of 2000 km and $700 for a flight of 4000 km.
(a) Plot these points on graph paper with distance on the horizontal axis and cost on
the vertical axis.
(b) Assuming a linear model, estimate
(i) the cost of a flight of 3200 km;
(ii) the distance travelled on a flight costing $400.
3. By substituting values into the equation, decide which of the following points lie on the
line, x + 4y = 12:
A(12, 0), B(2, 2), C(4, 2), D(−8, 5), E(0, 3)
4. For the line 3x − 5y = 8,
(a) find the value of x when y = 2;
(b) find the value of y when x = 1.
Hence write down the coordinates of two points which lie on this line.
5. If 4x + 3y = 24, complete the following table and hence sketch this line.
x
y
0
0
3
6. Solve the following pairs of simultaneous linear equations graphically:
(a) −2x + y = 2
(b) 3x + 4y = 12
x + 4y = 8
2x + y = −6 (c) 2x + y = 4 (d)
x+y=1
4x − 3y = 3 6x + 5y = 15
7. State the value of the slope and y intercept for each of the following lines:
(a) y = 5x + 9
(b) y = 3x − 1
(c) y = 13 − x
(d) −x + y = 4
(e) 4x + 2y = 5
(f) 5x − y = 6
8. Use the slope–intercept approach to produce a rough sketch of the following lines:
(a) y = −x
(b) x − 2y = 6
9. A taxi firm charges a fixed cost of $4 plus a charge of $2.50 a mile.
(a) Write down a formula for the cost, C, of a journey of x miles.
(b) Plot a graph of C against x for 0 ≤ x ≤ 20.
(c) Hence, or otherwise, work out the distance of a journey which costs $24.
Section 1.3 Graphs of linear equations 53
10. The number of people, N, employed in a chain of cafes is related to the number of cafes,
n, by the equation:
N = 10n + 120
(a) Illustrate this relation by plotting a graph of N against n for 0 ≤ n ≤ 20.
(b) Hence, or otherwise, calculate the number of
(i) employees when the company has 14 cafes;
(ii) cafes when the company employs 190 people.
(c) State the values of the slope and intercept of the graph and give an interpretation.
11. Monthly sales revenue, S (in $), and monthly advertising expenditure, A (in $), are
modelled by the linear relation, S = 9000 + 12A.
(a) If the firm does not spend any money on advertising, what is the expected sales
revenue that month?
(b) If the firm spends $800 on advertising one month, what is the expected sales revenue?
(c) How much does the firm need to spend on advertising to achieve monthly sales
revenue of $15 000?
(d) If the firm increases monthly expenditure on advertising by $1, what is the corres­
ponding increase in sales revenue?
Exercise 1.3*
1. Which of the following points lie on the line 3x − 5y = 25?
(5, −2), (10, 1), (−5, 0) (5, 10), (−5, 10), (0, −5)
2. Solve the following pairs of simultaneous equations graphically:
(a) y = 3x − 1
(b) 2x + y = 6
(c) 2x + 3y = 5
(d) 3x + 4y = −12
y = 2x + 1 x − y = −3 5x − 2y = −16 −2x + 3y = 25
3. State the value of the slope and y intercept for each of the following lines:
(a) y = 7x − 34
(b) y = 1 − x
(c) 3x − 2y = 6
(e) x − 5y = 0
(f) y = 2
(g) x = 4
(d) −4x + 2y = 5
4. Identify the two lines in the following list which are parallel:
(a) 3x + 5y = 2
(b) 5x − 3y = 1
(d) 10x − 6y = 9
(e) y = 0.6x + 2
(c) 5x + 3y = 13
5. (a) The Feelgood Gym has a monthly membership fee of $100 and it charges an addi-
tional $1 per hour. If I use the gym for x hours in a month, write down an expression
for the total cost in terms of x.
(b) Repeat part (a) for the FabulousMe Gym, which charges $35 per month and $3.50
per hour.
(c) By plotting both graphs on the same axes, or otherwise, find the number of hours
which gives the same total cost.
➜
54 Chapter 1 Linear Equations
6. A bakery discovers that if it decreases the price of its birthday cakes by $1, it sells 12
more cakes each month.
(a) Assuming that monthly sales, M, are related to prices, P, by a linear model,
M = aP + b, state the value of a.
(b) If the bakery sells 240 cakes in a month when the price of the cake is $14, work out
the value of b.
(c) Use this model to estimate monthly sales when the price is $9.
(d) If the bakery can make only 168 cakes in a month, work out the price that it needs
to charge to sell them all.
7. (1) Show that the lines ax + by = c and dx + ey = f are parallel whenever ae − bd = 0.
(2) Use the result of part (1) to comment on the solution of the following simultaneous
equations:
2x − 4y = 1
−3x + 6y = 7
8. Write down the coordinates of the points where the line ax + by = c intercepts the axes.
Section 1.4
Algebraic solution of simultaneous
linear equations
Objectives
At the end of this section you should be able to:
●
Solve a system of two simultaneous linear equations in two unknowns using elimination.
●Detect
when a system of equations does not have a solution.
●Detect
when a system of equations has infinitely many solutions.
●
Solve a system of three simultaneous linear equations in three unknowns using elimination.
In Section 1.3 a graphical method for the solution of simultaneous linear equations was
described. Both lines are sketched on the same piece of graph paper and the coordinates
of the point of intersection are then simply read off from the diagram. Unfortunately, this
approach has several drawbacks. It is not always easy to decide on a suitable scale for the
axes. Even if the scale allows all four points (two from each line) to fit on the diagram, there
is no guarantee that the point of intersection itself also lies on it. When this happens, you
have no alternative but to throw away your graph paper and to start again, choosing a
smaller scale in the hope that the solution will now fit. The second drawback concerns the
accuracy of the graphical solution. All of the problems in Section 1.3 were deliberately chosen
so that the answers had nice numbers in them; whole numbers such as −1, 2 and 5 or, at
worst, simple fractions such as 1/2, 21/2 and −1/4. In practice, the coefficients of the equations
may well involve decimals and we might expect a decimal solution. Indeed, even if the coefficients are whole numbers, the solution itself could involve nasty fractions such as 7/8 or
perhaps something like 231/571. A moment’s thought should convince you that in these circumstances it is virtually impossible to obtain the solution graphically, even if we use a really
large scale and our sharpest HB pencil in the process. The final drawback concerns the nature
of the problem itself. Quite frequently in economics, we need to solve three equations in three
unknowns or maybe four equations in four unknowns. Unfortunately, the graphical method
of solution does not extend to these cases.
In this section an alternative method of solution is described which relies on algebra. It is
called the elimination method, since each stage of the process eliminates one (or more) of the
unknowns. This method always produces the exact solution and can be applied to systems of
equations larger than just two equations in two unknowns. In order to illustrate the method,
we return to the simple example considered in the previous section:
4x + 3y = 11
(1)
2x + y = 5
(2)
56 Chapter 1 Linear Equations
The coefficient of x in equation (1) is 4 and the coefficient of x in equation (2) is 2. If these
numbers had turned out to be exactly the same, then we could have eliminated the variable
x by subtracting one equation from the other. However, we can arrange for this to be the case
by multiplying the left-hand side of the second equation by 2. Of course, we must also
remember to multiply the right-hand side of the second equation by 2 in order for this operation to be valid. The second equation then becomes
4x + 2y = 10
(3)
We may now subtract equation (3) from equation (1) to get
y=1
You may like to think of this in terms of the usual layout for the subtraction of two ordinary
numbers: that is,
4x 1 3y 5 11
4x 1 3y 5 10 2
y5 1
the xs cancel
when you subtract
This number can now be substituted into one of the original equations to deduce x. From
equation (1)
4x + 3(1) = 11 (substitute y = 1)
4x + 3 = 11
4x = 8
x= 2
(subtract 3 from both sides)
(divide both sides by 4)
Hence the solution is x = 2, y = 1. As a check, substitution of these values into the other
original equation (2) gives
2(2) + 1 = 5 ✓
The method of elimination can be summarised as follows.
Step 1
Add/subtract a multiple of one equation to/from a multiple of the other to eliminate x.
Step 2
Solve the resulting equation for y.
Step 3
Substitute the value of y into one of the original equations to deduce x.
Step 4
Check that no mistakes have been made by substituting both x and y into the other original
equation.
Section 1.4 Algebraic solution of simultaneous linear equations 57
Example
Solve the system of equations
3x + 2y = 1
(1)
−2x + y = 2
(2)
Solution
Step 1
The coefficients of x in equations (1) and (2) are 3 and −2, respectively. We can arrange
for these to be the same size (but of opposite sign) by multiplying equation (1) by 2 and
multiplying equation (2) by 3. The new equations will then have x coefficients of 6 and −6,
so we can eliminate x this time by adding the equations together. The details are as follows.
Doubling the first equation produces
6x + 4y = 2
(3)
Tripling the second equation produces
−6x + 3y = 6
(4)
If equation (4) is added to equation (3), then
6x 1 4y 5 2
26x 1 3y 5 6 1
7y 5 8
the xs cancel
when you add
(5)
Step 2
Equation (5) can be solved by dividing both sides by 7 to get
y = 8/7
Step 3
If 8/7 is substituted for y in equation (1), then
3x 1 2
3x 1
8
51
7
16
51
7
16
7
7 2 16
3x 5
7
3x 5 1 2
3x 5 2
x5
(put over a common denominator)
9
7
1
9
3 2
3
7
x 52
(subtract 16/7 from both sides)
(divide both sides by 3)
3
7
The solution is therefore x = −3/7, y = 8/7.
➜
58 Chapter 1 Linear Equations
Step 4
As a check, equation (2) gives
22 2
3
8 6 8 6 1 8 14
5 2 ✓
1 5 1 5
5
7
7 7 7
7
7
Advice
In the general description of the method, we suggested that the variable x is eliminated in step 1.
There is nothing special about x. We could equally well eliminate y at this stage and then solve the
resulting equation in step 2 for x.
You might like to solve the above example using this alternative strategy. You need to double
equation (2) and then subtract from equation (1).
Practice Problem
1. (a) Solve the equations
3x − 2y = 4
x − 2y = 2
by eliminating one of the variables.
(b) Solve the equations
3x + 5y = 19
−5x + 2y = −11
by eliminating one of the variables.
The following examples provide further practice in using the method and illustrate some
special cases which may occur.
Example
Solve the system of equations
x − 2y = 1
2x − 4y = −3
Solution
The variable x can be eliminated by doubling the first equation and subtracting the second:
both the xs and
the ys cancel!
2x 2 4y 5 2
2x 2 4y 5 23 2
05 5
Section 1.4 Algebraic solution of simultaneous linear equations 59
The statement ‘0 = 5’ is clearly nonsense and something has gone seriously wrong. To
understand what is going on here, let us try to solve this problem graphically.
The line x − 2y = 1 passes through the points (0, −1/2) and (1, 0) (check this). The line
2x − 4y = −3 passes through the points (0, 3/4) and (−3/2, 0) (check this). Figure 1.11
shows that these lines are parallel and so they do not intersect. It is therefore not surprising
that we were unable to find a solution using algebra, because this system of equations does
not have one. We could have deduced this before when subtracting the equations. The
equation that only involves y in step 2 can be written as
0y = 5
and the problem is to find a value of y for which this equation is true. No such value exists,
since
zero 3 any number 5 zero
and so the original system of equations does not have a solution.
y
1
2x ] 4y 5 ]3
1
2
]1
] 12
1
2
] 12
1
x
x ] 2y 5 1
]1
Figure 1.11
Example
Solve the equations
2x − 4y = 1
5x − 10y = 5/2
Solution
The variable x can be eliminated by multiplying the first equation by 5, multiplying the
second equation by 2 and subtracting
➜
60 Chapter 1 Linear Equations
10x 2 20y 5 5
10x 2 20y 5 5 2
050
everything cancels
including the
right-hand side!
Again, it is easy to explain this using graphs. The line 2x − 4y = 1 passes through (0, −1/4) and
(1/2, 0). The line 5x − 10y = 5/2 passes through (0, −1/4) and (1/2, 0). Consequently, both
equations represent the same line. From Figure 1.12 the lines intersect along the whole of their
length and any point on this line is a solution. This particular system of equations has infinitely
many solutions. This can also be deduced algebraically. The equation involving y in step 2 is
0y = 0
which is true for any value of y.
y
1
1
2
5x ] 10y 5 5
2
2x ] 4y 5 1
1
2
]
1
x
1
2
]1
Figure 1.12
Practice Problem
2. Attempt to solve the following systems of equations:
(a)
3x − 6y = −2
−4x + 8y = −1
(b) −5x + y = 4
10x − 2y = −8
Comment on the nature of the solution in each case.
It is possible to identify when simultaneous equations fail to possess a unique solution by
considering the general case:
ax + by = c
dx + ey = f
The variable y can be eliminated by multiplying the first equation by e, multiplying the second equation by b and subtracting:
aex 1 bey 5 ce
bdx 1 bey 5 bf
2
(ae 2 bd)x 5 ce 2 bf
Section 1.4 Algebraic solution of simultaneous linear equations 61
so that
x5
ce 2 bf
ae 2 bd
In the same way it is possible to show that
y5
af 2 cd
(see Exercise 1.4*, Question 3).
ae 2 bd
These formulae cannot be used when ae = bd since it is impossible to divide by zero. In
this case the system either has no solution or infinitely many solutions. You might like to
check that this condition holds for both of the systems in Practice Problem 2.
We now show how the algebraic method can be used to solve three equations in three
unknowns. As you might expect, the details are more complicated than for just two equations,
but the principle is the same. We begin with a simple example to illustrate the general method.
Consider the system
x + 3y − z = 4
(1)
2x + y + 2z = 10
(2)
3x − y + z = 4
(3)
The objective is to find three numbers x, y and z which satisfy these equations simultaneously. Our previous work suggests that we should begin by eliminating x from all but one of
the equations.
The variable x can be eliminated from the second equation by multiplying equation (1) by
2 and subtracting equation (2):
2x 1 6y 2 2z 5 8
2x 1 y 1 2z 5 10 2
5y 2 4z 5 22
(4)
Similarly, we can eliminate x from the third equation by multiplying equation (1) by 3 and
subtracting equation (3):
3x 1 9y 2 3z 5 12
3x 2 y 1 z 5 4 2
10y 2 4z 5 8
(5)
At this stage the first equation is unaltered but the second and third equations of the system
have changed to equations (4) and (5), respectively, so the current equations are
x + 3y − z =
4
(1)
5y − 4z = −2
(4)
10y − 4z =
8
(5)
Notice that the last two equations constitute a system of just two equations in two unknowns,
y and z. This, of course, is precisely the type of problem that we already know how to solve.
Once y and z have been calculated, the values can be substituted into equation (1) to deduce x.
We can eliminate y in the last equation by multiplying equation (4) by 2 and subtracting
equation (5):
10y 2 8z 5 24
10y 2 4z 5 8 2
24z 5 212
(6)
62 Chapter 1 Linear Equations
Collecting together the current equations gives
x + 3y − z =
4
(1)
5y − 4z = −2
(4)
−4z = −12
(6)
From the last equation,
z5
212
5 3 (divide both sides by −4)
24
If this is substituted into equation (4), then
5y − 4(3) = −2
5y − 12 = −2
5y = 10
y=
2
(add 12 to both sides)
(divide both sides by 5)
Finally, substituting y = 2 and z = 3 into equation (1) produces
x + 3(2) − 3 = 4
x+3=4
x = 1 (subtract 3 from sides)
Hence the solution is x = 1, y = 2, z = 3.
As usual, it is possible to check the answer by putting these numbers back into the original
equations (1), (2) and (3):
1 + 3(2) − 3 = 4 ✓
2(1) + 2 + 2(3) = 10 ✓
3(1) − 2 + 3 = 4 ✓
The general strategy may be summarised as follows. Consider the system
?x + ?y + ?z = ?
?x + ?y + ?z = ?
?x + ?y + ?z = ?
where ? denotes some numerical coefficient.
Step 1
Add/subtract multiples of the first equation to/from multiples of the second and third equations to eliminate x. This produces a new system of the form
?x + ?y + ?z = ?
?y + ?z = ?
?y + ?z = ?
Step 2
Add/subtract a multiple of the second equation to/from a multiple of the third to eliminate
y. This produces a new system of the form
?x + ?y + ?z = ?
?y + ?z = ?
?z = ?
Section 1.4 Algebraic solution of simultaneous linear equations 63
Step 3
Solve the last equation for z. Substitute the value of z into the second equation to deduce y.
Finally, substitute the values of both y and z into the first equation to deduce x.
Step 4
Check that no mistakes have been made by substituting the values of x, y and z into the
original equations.
It is possible to adopt different strategies from that suggested here. For example, it may be
more convenient to eliminate z from the last equation in step 2 rather than y. However, it is
important to notice that we use the second equation to do this, not the first. Any attempt to
use the first equation in step 2 would reintroduce the variable x into the equations, which is
the last thing we want to do at this stage.
Example
Solve the equations
4x + y + 3z = 8
(1)
−2x + 5y + z = 4
(2)
3x + 2y + 4z = 9
(3)
Solution
Step 1
To eliminate x from the second equation we multiply it by 2 and add to equation (1):
4x 1 y 1 3z 5 8
24x 1 10y 1 2z 5 8 1
11y 1 5z 5 16
(4)
To eliminate x from the third equation we multiply equation (1) by 3, multiply equation (3)
by 4 and subtract:
12x 1 3y 1 9z 5 24
12x 1 8y 1 16z 5 36 2
25y 2 7z 5 212
(5)
This produces a new system:
4x + y + 3z =    8
(1)
11y + 5z =   16
(4)
−5y − 7z = −12
(5)
Step 2
To eliminate y from the new third equation (that is, equation (5)) we multiply equation (4)
by 5, multiply equation (5) by 11 and add:
55y 1 25z 5
80
255y 2 77z 5 2132 1
252z 5 252
(6)
➜
64 Chapter 1 Linear Equations
This produces a new system
4x + y + 3z =
8
(1)
11y + 5z =
16
(4)
−52z = −52
(6)
Step 3
The last equation gives
252
5 1 (divide both sides by −52)
252
z5
If this is substituted into equation (4), then
11y + 5(1) = 16
11y + 5 = 16
11y = 11
y= 1
(subtract 5 from both sides)
(divide both sides by 11)
Finally, substituting y = 1 and z = 1 into equation (1) produces
4x + 1 + 3(1) = 8
4x + 4 = 8
4x = 4 (subtract 4 from both sides)
x = 1 (divide both sides by 4)
Hence the solution is x = 1, y = 1, z = 1.
Step 4
As a check the original equations (1), (2) and (3) give, respectively,
4(1) + 1 + 3(1) = 8 ✓
−2(1) + 5(1) + 1 = 4 ✓
3(1) + 2(1) + 4(1) = 9 ✓
Practice Problem
3. Solve the following system of equations:
2x + 2y − 5z = −5
(1)
x− y+ z= 3
(2)
−3x + y + 2z = −2
(3)
As you might expect, it is possible for three simultaneous linear equations to have
either no solution or infinitely many solutions. An illustration of this is given in Question 5
of Exercise 1.4*. The method described in this section has an obvious extension to larger
systems of equations. However, the calculations are extremely tedious to perform by hand.
Fortunately, many computer packages are available which can solve large systems accurately and efficiently (a matter of a few seconds to solve 10 000 equations in 10 000
unknowns).
Section 1.4 Algebraic solution of simultaneous linear equations 65
Advice
We shall return to the solution of simultaneous linear equations in Chapter 7 when we describe how
matrix theory can be used to solve them. This does not depend on any subsequent chapters in this
text, so you might like to read through this material now. Two techniques are suggested. A method
based on inverse matrices is covered in Section 7.2 and an alternative using Cramer’s rule can be
found in Section 7.3.
Key Term
Elimination method The method in which variables are removed from a system of simultaneous
equations by adding (or subtracting) a multiple of one equation to (or from) a multiple of another.
Exercise 1.4
1. Use the elimination method to solve the following pairs of simultaneous linear equations:
(a) −2x + y = 2
(b) 3x + 4y = 12
(c) 2x + y = 4
2x + y = −6 x + 4y = 8
(d) x + y = 1
4x − 3y = 3 6x + 5y = 15
2. The total annual sales of a book in either paper or electronic form are 3500. Each paper
copy of the book costs $30 and each e-book costs $25. The total cost is $97 500.
(a) If x and y denote the number of copies in paper and electronic form, write down a
pair of simultaneous equations.
(b) Solve the equations to find the number of e-books sold.
3. Sketch the following lines on the same diagram:
2x − 3y = 6, 4x − 6y = 18,
x2
3
y53
2
Hence comment on the nature of the solutions of the following systems of equations:
(a) 2x − 3y = 6
x2
(b) 4x − 6y = 18
3
3
y53
x 2 2 y 5 3
2
4. Use the elimination method to attempt to solve the following systems of equations.
Comment on the nature of the solution in each case.
(a) −3x + 5y = 4
(b) 6x − 2y = 3
9x − 15y = −12 15x − 5y = 4
5. If the following system of linear equations has infinitely many solutions, find the value
of k.
6x − 4y = 2
−3x + 2y = k
66 Chapter 1 Linear Equations
Exercise 1.4*
1. Solve the following pairs of simultaneous equations:
(a) y = 3x − 1 (b) 2x + y = 6
y = 2x + 1
x − y = −3
(c) 2x + 3y = 5
5x − 2y = −16
(d) 3x + 4y = −12
−2x + 3y = 25
2. Write down a possible set of values of the numbers a and b for which the simultaneous
equations:
(a) 2x + 3y = 4
have infinitely many solutions
ax + 6y = b
(b) 4x − 6y = 1
have no solutions
2x + ay = b
3. By eliminating x from the system
ax + by = c
dx + ey = f
show that
y5
af 2 cd
ae 2 bd
4. Solve the following systems of equations:
(a) x − 3y + 4z = 5 (1)
(b) 3x + 2y − 2z = −5 (1)
2x + y + z = 3 (2) 4x + 3y + 3z = 17 (2)
4x + 3y + 5z = 1 (3) 2x − y + z = −1 (3)
5. Attempt to solve the following systems of equations. Comment on the nature of the
sol­ution in each case.
(a) x − 2y + z = −2 (1)
x + y − 2z = 4
(2)
−2x + y + z = 12 (3)
(b) 2x + 3y − z = 13
(1)
x − 2y + 2z = −3 (2)
3x + y + z = 10
(3)
6. If the following system of equations has infinitely many solutions, find the value of the
constant, k.
x + 2y − 5z = 1
2x − y + 3z = 4
4x + 3y − 7z = k
What can you say about the nature of the solution for other values of k?
7. A distribution centre sends three different types of parcels. One consignment has
6 small, 8 medium and 9 large parcels which cost $173.20 to post. Another consignment
has 7 small, 13 medium and 17 large parcels with total postage $291.05. A large parcel
costs twice as much to post as a small one. Work out the total cost of posting 3 small
parcels, 9 medium parcels and 2 large parcels.
Section 1.5
Supply and demand analysis
Objectives
At the end of this section you should be able to:
●
Use the function notation, y = f(x).
●
Identify the endogenous and exogenous variables in an economic model.
●
Identify and sketch a linear demand function.
●
Identify and sketch a linear supply function.
●Determine
the equilibrium price and quantity for a single-commodity market both graphically
and algebraically.
●Determine
the equilibrium price and quantity for a multicommodity market by solving
simultaneous linear equations.
Microeconomics is concerned with the analysis of the economic theory and policy of individual firms and markets. In this section we focus on one particular aspect known as market
equilibrium, in which the supply and demand balance. We describe how the mathematics
introduced in the previous two sections can be used to calculate the equilibrium price and
quantity. However, before we do this it is useful to explain the concept of a function. This
idea is central to nearly all applications of mathematics in economics.
A function, f, is a rule which assigns to each incoming number, x, a uniquely defined out­
going number, y. A function may be thought of as a ‘black box’ that performs a dedicated
arithmetic calculation. As an example, consider the rule ‘double and add 3’. The effect of this
rule on two specific incoming numbers, 5 and −17, is illustrated in Figure 1.13.
Unfortunately, such a representation is rather cumbersome. There are, however, two
alternative ways of expressing this rule which are more concise. We can write either
y = 2x + 3 or f(x) = 2x + 3
The first of these is familiar to you from our previous work; corresponding to any incoming
number, x, the right-hand side tells you what to do with x to generate the outgoing number, y.
5
double and
add 3
13
IN
OUT
]17
Figure 1.13
double and
add 3
]31
68 Chapter 1 Linear Equations
The second notation is also useful. It has the advantage that it involves the label f, which is
used to name the rule. If, in a piece of economic theory, there are two or more functions, we
can use different labels to refer to each one. For example, a second function might be
g(x) = −3x + 10
and we subsequently identify the respective functions simply by referring to them by name:
that is, as either f or g.
The new notation also enables the information conveyed in Figure 1.13 to be written
f(5) = 13 read ‘f of 5 equals 13’
f(−17) = −31 read ‘f of ]17 equals ]31’
The number inside the brackets is the incoming value, x, and the right-hand side is the
corresponding outgoing value, y.
Practice Problem
1. Evaluate
(a) f(25) (b) f (1) (c) f (17) (d) g(0) (e) g(48) (f) g(16)
for the two functions
f (x) = −2x + 50
g(x) = −1/2 x + 25
Do you notice any connection between f and g?
The incoming and outgoing variables are referred to as the independent and dependent
variables, respectively. The value of y clearly ‘depends’ on the actual value of x that is fed
into the function. For example, in microeconomics the quantity demanded, Q, of a good
depends on the market price, P. We might express this as
Q = f (P)
Such a function is called a demand function. Given any particular formula for f(P), it is then
a simple matter to produce a picture of the corresponding demand curve on graph paper.
There is, however, a difference of opinion between mathematicians and economists on how
this should be done. If your quantitative methods lecturer is a mathematician, then that person is likely to plot Q on the vertical axis and P on the horizontal axis. Economists, on the
other hand, normally plot them the other way round, with Q on the horizontal axis. In doing
so, we are merely noting that since Q is related to P, then, conversely, P must be related to
Q, and so there is a function of the form
P = g(Q)
The two functions, f and g, are said to be inverse functions: that is, f is the inverse of g
and, equivalently, g is the inverse of f. We adopt the economists’ approach in this text. In
Section 1.5 Supply and demand analysis 69
sub­sequent chapters we shall investigate other microeconomic functions such as total revenue,
average cost and profit. It is conventional to plot each of these against Q (that is, with Q on
the horizontal axis), so it makes sense to be consistent and to do the same here.
Written in the form P = g(Q), the demand function tells us that P is a function of Q but
gives us no information about the precise relationship between these two variables. To find
this we need to know the form of the function which can be obtained either from economic
theory or from empirical evidence. For the moment we hypothesise that the function is linear,
so that
P = aQ + b
for some appropriate constants (called parameters), a and b. Of course, in reality, the relationship between price and quantity is likely to be much more complicated than this. However,
the use of linear functions makes the mathematics nice and easy, and the result of any analysis
at least provides a first approximation to the truth. The process of identifying the key features
of the real world and making appropriate simplifications and assumptions is known as
modelling. Models are based on economic laws and help to explain and predict the behaviour
of real-world situations. Inevitably, there is a conflict between mathematical ease and the model’s
accuracy. The closer the model comes to reality, the more complicated the mathematics is
likely to be.
A graph of a typical linear demand function is shown in Figure 1.14. Elementary theory
shows that demand usually falls as the price of a good rises and so the slope of the line is
negative. Mathematically, P is then said to be a decreasing function of Q.
In symbols we write
a < 0 read ‘a is
less than zero’
It is also apparent from the graph that the intercept, b, is positive: that is,
b > 0 read ‘b is
greater than zero’
In fact, it is possible in theory for the demand curve to be horizontal with a = 0. This corres­
ponds to perfect competition, and we shall return to this special case in Chapter 4.
P
b
P 5 aQ 1 b
Q
Figure 1.14
70 Chapter 1 Linear Equations
Example
Sketch a graph of the demand function
P = −2Q + 50
Hence, or otherwise, determine the value of
(a) P when Q = 9
(b) Q when P = 10
Solution
For the demand function
P = −2Q + 50
a = −2, b = 50, so the line has a slope of −2 and an intercept of 50. For every 1 unit along,
the line goes down by 2 units, so it must cross the horizontal axis when Q = 25.
(Alternatively, note that when P = 0 the equation reads 0 = −2Q + 50, with solution
Q = 25.) The graph is sketched in Figure 1.15.
(a) Given any quantity, Q, it is straightforward to use the graph to find the corresponding
price, P. A line is drawn vertically upwards until it intersects the demand curve and the
value of P is read off from the vertical axis. From Figure 1.15, when Q = 9 we see that
P = 32. This can also be found by substituting Q = 9 directly into the demand function to get
P = −2(9) + 50 = 32
(b) Reversing this process enables us to calculate Q from a given value of P. A line is drawn
horizontally until it intersects the demand curve, and the value of Q is read off from
the horizontal axis. Figure 1.15 indicates that Q = 20 when P = 10. Again, this can be
found by calculation. If P = 10, then the equation reads
10 = −2Q + 50
−40 = −2Q
20 = Q
(subtract 50 from both sides)
(divide both sides by −2)
P
50
32
P 5 ]2Q 1 50
10
9
Figure 1.15
20
25
Q
Section 1.5 Supply and demand analysis 71
Practice Problem
2. Sketch a graph of the demand function
P = −3Q + 75
Hence, or otherwise, determine the value of
(a) P when Q = 23
(b) Q when P = 18
Example
A potter makes and sells ceramic bowls. It is observed that when the price is $32, only
9 bowls are sold in a week; but when the price decreases to $10, weekly sales rise to 20.
Assuming that demand can be modelled by a linear function,
(a) obtain a formula for P in terms of Q;
(b) sketch a graph of P against Q;
(c) comment on the likely reliability of the model.
Solution
(a) The general formula for a linear demand function is P = aQ + b, where a is the slope
of the line. The values of the parameters a and b can be worked out in two different ways.
Method 1 – Calculate the slope of the line
Notice that weekly sales have gone up by 11 bowls as a result of a $22 decrease in price. If
the relationship is linear, then a 1-unit increase in Q corresponds to a 22/11 = 2-unit decrease
in P, so the graph has a slope of −2. The equation must be given by P = −2Q + b. To find
b we can use the fact that when Q = 9, P = 32, so 32 = −2 × 9 + b, giving b = 50. The
demand function is
P = −2Q + 50
Method 2 – Simultaneous equations
When Q = 9, we know that P = 32, so 9a + b = 32
When Q = 20, we know that P = 10, so 20a + b = 10
Subtracting the second equation from the first gives −11a = 22, so a = −2. This value
can be substituted into either equation to get b = 50 as before.
(b) The graph is sketched in Figure 1.16 using the fact that the intercept is 50 and the slope
is −2. Alternatively, it could be sketched using the fact that it passes through (9, 32)
and (20, 10).
(c) Although the assumption of a linear relationship may be valid in the middle of the range,
it is unlikely to be true at the ends. If the potter were to give away the bowls for free, it
is unlikely that the demand would be limited to 25. Similarly, at the other end it would
be surprising if no one buys a bowl the moment the price goes above $50. In practice the
graph will be curved at each end as shown by the dashed lines in Figure 1.16.
➜
72 Chapter 1 Linear Equations
P
50
32
10
9
20
25
Q
Figure 1.16
The model of consumer demand given so far is fairly crude in that it assumes that quantity
depends solely on the price, P, of the good being considered. In practice, Q depends on other
factors as well. These include the incomes of consumers, Y, the price of substitutable goods,
PS, the price of complementary goods, PC, advertising expenditure, A, and consumers’ tastes, T.
A substitutable good is one that could be consumed instead of the good under consideration.
For example, in the transport industry, buses and taxis could obviously be substituted for
each other in urban areas. A complementary good is one that is used in conjunction with other
goods. For example, laptops and printers are consumed together. Mathematically, we say
that Q is a function of P, Y, PS, PC, A and T. This is written as
Q = f (P, Y, PS, PC, A, T)
where the variables inside the brackets are separated by commas. In terms of a ‘black box’
diagram, this is represented with six incoming lines and one outgoing line as shown in
Figure 1.17. In our previous discussion it was implicitly assumed that the variables Y, PS, PC,
A and T are held fixed. We describe this situation by calling Q and P endogenous variables,
since they are allowed to vary and are determined within the model. The remaining variables
are called exogenous, since they are constant and are determined outside the model.
P
Y
Ps
Pc
A
T
Figure 1.17
f
Q
Section 1.5 Supply and demand analysis 73
P
G
E
P*
Q1
F
Q2
H
Q
Figure 1.18
Let us return now to the standard demand curve shown in Figure 1.18 as the line EF. This
is constructed on the assumption that Y, PS, PC, A and T are all constant. Notice that when the
price is P* the quantity demanded is Q1. Now suppose that income, Y, increases. We would
normally expect the demand to rise because the extra income buys more goods at price P*.
The effect is to shift the demand curve to the right because at price P* consumers can afford
the larger number of goods, Q2. From Figure 1.18 we deduce that if the demand curve is
P = aQ + b
then a rise in income causes the intercept, b, to increase.
We conclude that if one of the exogenous variables changes, then the whole demand curve
moves, whereas if one of the endogenous variables changes, we simply move along the fixed
curve.
Incidentally, it is possible that, for some goods, an increase in income actually causes the
demand curve to shift to the left. In the 1960s and 1970s, most western economies saw a
decline in the domestic consumption of coal as a result of an increase in income. In this case,
higher wealth meant that more people were able to install central heating systems which use
alternative forms of energy. Under these circumstances the good is referred to as an inferior
good. On the other hand, a normal good is one whose demand rises as income rises. Cars
and electrical goods are obvious examples of normal goods. Currently, concern about global
warming is also reducing demand for coal. This factor can be incorporated as part of taste,
although it is difficult to handle mathematically since it is virtually impossible to quantify
taste and so to define T numerically.
The supply function is the relation between the quantity, Q, of a good that producers
plan to bring to the market and the price, P, of the good. A typical linear supply curve is
indicated in Figure 1.19. Economic theory indicates that, as the price rises, so does the supply.
Mathematically, P is then said to be an increasing function of Q. A price increase encourages
existing producers to raise output and entices new firms to enter the market. The line shown
in Figure 1.19 has equation
P = aQ + b
74 Chapter 1 Linear Equations
P
P 5 aQ 1 b
b
Q
Figure 1.19
with slope a > 0 and intercept b > 0. Note that when the market price is equal to b, the supply is zero. Only when the price exceeds this threshold level do producers decide that it is
worth supplying any good whatsoever.
Again, this is a simplification of what happens in the real world. The supply function does
not have to be linear and the quantity supplied, Q, is influenced by things other than price.
These exogenous variables include the prices of factors of production (that is, land, capital,
labour and enterprise), the profits obtainable on alternative goods, and technology.
In microeconomics we are concerned with the interaction of supply and demand.
Figure 1.20 shows typical supply and demand curves sketched on the same diagram. Of
particular significance is the point of intersection. At this point the market is in equilibrium
because the quantity supplied exactly matches the quantity demanded. The corresponding
price, P0, and quantity, Q0, are called the equilibrium price and quantity.
In practice, it is often the deviation of the market price away from the equilibrium price
that is of most interest. Suppose that the market price, P*, exceeds the equilibrium price, P0.
From Figure 1.20 the quantity supplied, QS, is greater than the quantity demanded, QD, so
there is excess supply. There are stocks of unsold goods, which tend to depress prices and
cause firms to cut back production. The effect is for ‘market forces’ to shift the market back
P
supply curve
P*
P0
demand
curve
QD Q0
Figure 1.20
QS
Section 1.5 Supply and demand analysis 75
down towards equilibrium. Likewise, if the market price falls below equilibrium price, then
demand exceeds supply. This shortage pushes prices up and encourages firms to produce
more goods, so the market drifts back up towards equilibrium.
Example
The demand and supply functions of a good are given by
P = −2QD + 50
P = 1/ 2 QS + 25
where P, QD and QS denote the price, quantity demanded and quantity supplied, respectively.
(a) Determine the equilibrium price and quantity.
(b) Determine the effect on the market equilibrium if the government decides to impose a
fixed tax of $5 on each good.
Solution
(a) The demand curve has already been sketched in Figure 1.15. For the supply function
P = 1/2 QS + 25
we have a = 1/ 2, b = 25, so the line has a slope of 1/ 2 and an intercept of 25. It therefore
passes through (0, 25). For a second point, let us choose QS = 20, say. The corresponding value of P is
P = 1/2(20) + 25 = 35
so the line also passes through (20, 35). The points (0, 25) and (20, 35) can now be plotted
and the supply curve sketched. Figure 1.21 shows both the demand and supply curves
sketched on the same diagram. The point of intersection has coordinates (10, 30), so
the equilibrium quantity is 10 and the equilibrium price is 30.
P
50
supply curve
after tax
supply curve
before tax
(20, 35)
30
25
point of intersection
is (10, 30)
demand curve
25
Figure 1.21
Q
➜
76 Chapter 1 Linear Equations
It is possible to calculate these values using algebra. In equilibrium, QD = QS. If this
common value is denoted by Q, then the demand and supply equations become
P = −2Q + 50
and P = 1/ 2Q + 25
This represents a pair of simultaneous equations for the two unknowns P and Q
and so could be solved using the elimination method described in the previous section.
However, this is not strictly necessary because it follows immediately from the above
equations that
−2Q + 50 = 1/ 2 Q + 25
since both sides are equal to P. This can be rearranged to calculate Q:
−21/ 2 Q + 50 = 25
(subtract 1/2 Q from both sides)
−21/ 2 Q = −25 (subtract 50 from both sides)
Q = 10
(divide both sides by −21/2 )
Finally, P can be found by substituting this value into either of the original equations.
The demand equation gives
P = −2(10) + 50 = 30
As a check, the supply equation gives
P = 1/2(10) + 25 = 30 ✓
(b) If the government imposes a fixed tax of $5 per good, then the money that the firm
actually receives from the sale of each good is the amount, P, that the consumer pays,
less the tax, 5: that is, P − 5. Mathematically, this problem can be solved by replacing
P with P − 5 in the supply equation to get the new supply equation
P − 5 = 1/2 QS + 25
that is,
P = 1/2 QS + 30
The remaining calculations proceed as before. In equilibrium, QD = QS. Again, setting
this common value to be Q gives
P = −2Q + 50
P = 1/2 Q + 30
Hence
−2Q + 50 = 1/2 Q + 30
which can be solved as before to give Q = 8. Substitution into either of the above equations gives P = 34. (Check the details.)
Graphically, the introduction of tax shifts the supply curve upwards by 5 units.
Obviously, the demand curve is unaltered. The dashed line in Figure 1.21 shows the new
supply curve, from which the new equilibrium quantity is 8 and equilibrium price is 34.
Note the effect that government taxation has on the market equilibrium price. This
has risen to $34 and so not all of the tax is passed on to the consumer. The consumer
pays an additional $4 per good. The remaining $1 of tax must, therefore, be paid by
the firm.
Section 1.5 Supply and demand analysis 77
Practice Problem
3. The demand and supply functions of a good are given by
P = −4QD + 120
P = 1/3 QS + 29
where P, QD and QS denote the price, quantity demanded and quantity supplied, respectively.
(a) Calculate the equilibrium price and quantity.
(b) Calculate the new equilibrium price and quantity after the imposition of a fixed tax
of $13 per good. Who pays the tax?
We conclude this section by considering a more realistic model of supply and demand,
taking into account substitutable and complementary goods. Let us suppose that there are
two goods in related markets, which we call good 1 and good 2. The demand for either good
depends on the prices of both good 1 and good 2. If the corresponding demand functions are
linear, then
QD1 = a1 + b1 P1 + c1P2
QD2 = a2 + b2 P1 + c2 P2
where Pi and QDi denote the price and demand for the ith good and ai, bi and ci are parameters. For the first equation, a1 > 0 because there is a positive demand when the prices of both
goods are zero. Also, b1 < 0 because the demand of a good falls as its price rises. The sign of
c1 depends on the nature of the goods. If the goods are substitutable, then an increase in the
price of good 2 would mean that consumers would switch from good 2 to good 1, causing
QD1 to increase. Substitutable goods are therefore characterised by a positive value of c1.
On the other hand, if the goods are complementary, then a rise in the price of either good
would see the demand fall, so c1 is negative. Similar results apply to the signs of a2, b2 and c2.
The calculation of the equilibrium price and quantity in a two-commodity market model is
demonstrated in the following example.
Example
The demand and supply functions for two interdependent commodities are given by
QD1 = 10 − 2P1 + P2
QD2 = 5 + 2P1 − 2P2
QS1 = −3 + 2P1
QS2 = −2 + 3P2
where QDi, QSi and Pi denote the quantity demanded, quantity supplied and price of good i,
respectively. Determine the equilibrium price and quantity for this two-commodity model.
Solution
In equilibrium, we know that the quantity supplied is equal to the quantity demanded for
each good, so that
QD1 = QS1 and QD2 = QS2
➜
78 Chapter 1 Linear Equations
Let us write these respective common values as Q1 and Q2. The demand and supply equations
for good 1 then become
Q1 = 10 − 2P1 + P2
Q1 = −3 + 2P1
Hence
10 − 2P1 + P2 = −3 + 2P1
since both sides are equal to Q1. It makes sense to tidy this equation up a bit by collecting
all of the unknowns on the left-hand side and putting the constant terms on to the righthand side:
(subtract 2P1 from both sides)
10 − 4P1 + P2 = −3
−4P1 + P2 = −13
(subtract 10 from both sides)
We can perform a similar process for good 2. The demand and supply equations become
Q2 = 5 + 2P1 − 2P2
Q2 = −2 + 3P2
because QD2 = QS2 = Q2 in equilibrium. Hence
5 + 2P1 − 2P2 = −2 + 3P2
(subtract 3P2 from both sides)
5 + 2P1 − 5P2 = −2
(subtract 5 from both sides)
2P1 − 5P2 = −7
We have therefore shown that the equilibrium prices, P1 and P2, satisfy the simultaneous
linear equations
−4P1 + P2 = −13
(1)
2P1 − 5P2 = −7
(2)
which can be solved by elimination. Following the steps described in Section 1.4, we proceed as follows.
Step 1
Double equation (2) and add to equation (1) to get
24P1 1 P2 5 213
4P1 2 10P2 5 214 1
29P2 5 227
Step 2
Divide both sides of equation (3) by −9 to get P2 = 3.
Step 3
If this is substituted into equation (1), then
−4P1 + 3 = −13
−4P1 = −16
P1 = 4
(subtract 3 from both sides)
(divide both sides by −4)
(3)
Section 1.5 Supply and demand analysis 79
Step 4
As a check, equation (2) gives
2(4) − 5(3) = −7 ✓
Hence P1 = 4 and P2 = 3.
Finally, the equilibrium quantities can be deduced by substituting these values back into
the original supply equations. For good 1,
Q1 = −3 + 2P1 = −3 + 2(4) = 5
For good 2,
Q2 = −2 + 3P2 = −2 + 3(3) = 7
As a check, the demand equations also give
Q1 = 10 − 2P1 + P2 = 10 − 2(4) + 3 = 5 ✓
Q2 = 5 + 2P1 − 2P2 = 5 + 2(4) − 2(3) = 7 ✓
Practice Problem
4. The demand and supply functions for two interdependent commodities are given by
QD1 = 40 − 5P1 − P2
QD2 = 50 − 2P1 − 4P2
QS1 = −3 + 4P1
QS2 = −7 + 3P2
where QDi, QSi and Pi denote the quantity demanded, quantity supplied and price of good
i, respectively. Determine the equilibrium price and quantity for this two-commodity
model. Are these goods substitutable or complementary?
For a two-commodity market the equilibrium prices and quantities can be found by
solving a system of two simultaneous equations. Exactly the same procedure can be applied
to a three-commodity market, which requires the solution of a system of three simultaneous
equations.
Advice
An example of a three-commodity model can be found in Question 6 of Exercise 1.5*. Alternative
methods and further examples are described in Chapter 7. In general, with n goods it is necessary
to solve n equations in n unknowns and, as pointed out in Section 1.4, this is best done using a
computer package whenever n is large.
80 Chapter 1 Linear Equations
Key Terms
Complementary goods A pair of goods consumed together. As the price of either goes up, the
demand for both goods goes down.
Decreasing function
A function, y = f(x), in which y decreases as x increases.
Demand function A relationship between the quantity demanded and various factors that affect
demand, including price.
Dependent variable A variable whose value is determined by that taken by the independent variables; in y = f(x), the dependent variable is y.
Endogenous variable
A variable whose value is determined within a model.
Equilibrium (market) This state occurs when quantity supplied and quantity demanded are equal.
Exogenous variable A variable whose value is determined outside a model.
Function A rule that assigns to each incoming number, x, a uniquely defined outgoing number, y.
Increasing function
A function, y = f(x), in which y increases as x increases.
Independent variable A variable whose value determines that of the dependent variable; in
y = f(x), the independent variable is x.
Inferior good
A good whose demand decreases as income increases.
Inverse function A function, written f −1, which reverses the effect of a given function, f, so that
x = f −1(y) when y = f(x).
Modelling The creation of a piece of mathematical theory which represents (a simplification of)
some aspect of practical economics.
Normal good
A good whose demand increases as income increases.
Parameter A constant whose value affects the specific values but not the general form of a
mathematical expression, such as the constants a, b and c in ax2 + bx + c.
Substitutable goods A pair of goods that are alternatives to each other. As the price of one good
goes up, the demand for the other rises.
Supply function A relationship between the quantity supplied and various factors that affect
supply, including price.
Exercise 1.5
1. If f (x) = 3x + 15 and g(x) = 1/3x − 5, evaluate
(a) f(2) (b) f(10) (c) f(0) (d) g(21) (e) g(45) (f) g(15)
What word describes the relationship between f and g?
2. Sketch a graph of the supply function
P = 1/3 Q + 7
Hence, or otherwise, determine the value of
(a) P when Q = 12
(b) Q when P = 10
(c) Q when P = 4
Section 1.5 Supply and demand analysis 81
3. The demand function of a good is
Q = 100 − P + 2Y + 1/2 A
where Q, P, Y and A denote quantity demanded, price, income and advertising expenditure, respectively.
(a) Calculate the demand when P = 10, Y = 40 and A = 6. Assuming that price and
income are fixed, calculate the additional advertising expenditure needed to raise
demand to 179 units.
(b) Is this good inferior or normal?
4. The demand, Q, for a certain good depends on its own price, P, and the price of an
alterna­tive good, PA, according to
Q = 30 − 3P + PA
(a) Find Q if P = 4 and PA = 5.
(b) Is the alternative good substitutable or complementary? Give a reason for your answer.
(c) Determine the value of P if Q = 23 and PA = 11.
5. The demand for a good priced at $50 is 420 units, and when the price is $80, demand
is 240 units. Assuming that the demand function takes the form Q = aP + b, find the
values of a and b.
6. (a) Copy and complete the following table of values for the supply function
P = 1/2 Q + 20
Q
0
P
50
25
Hence, or otherwise, draw an accurate sketch of this function using axes with values
of Q and P between 0 and 50.
(b) On the same axes draw the graph of the demand function
P = 50 − Q
and hence find the equilibrium quantity and price.
(c) The good under consideration is normal. Describe the effect on the equilibrium
quantity and price when income rises.
7. The demand and supply functions of a good are given by
P = −3QD + 48
P = 1/2 QS + 23
Find the equilibrium quantity if the government imposes a fixed tax of $4 on each good.
8. The demand and supply functions for two interdependent commodities are given by
QD1 = 100 − 2P1 + P2
QD2 = 5 + 2P1 − 3P2
QS1 = −10 + P1
QS2 = −5 + 6P2
➜
82 Chapter 1 Linear Equations
where QDi, QSi and Pi denote the quantity demanded, quantity supplied and price of good
i, respectively. Determine the equilibrium price and quantity for this two-commodity model.
9. A demand function of a certain good is given by
Q = −20P + 0.04Y + 4T + 3Pr
where Q and P denote the quantity and price of the good, Y is income, T is taste and
Pr is the price of a related good.
(a) Calculate Q when P = 8, Y = 1000, T = 15 and Pr = 30.
(b) Is the related good substitutable or complementary? Give a reason for your answer.
(c) Find the value of P when Q = 235, Y = 8000, T = 30 and Pr = 25.
(d) The exogenous variables are now fixed at Y = 2000, T = 10 and Pr = 5. State the
values of the slope and vertical intercept when the demand function is sketched with
(i) P on the horizontal axis and Q on the vertical axis;
(ii) Q on the horizontal axis and P on the vertical axis.
Exercise 1.5*
1. Describe the effect on the demand curve due to an increase in
(a) the price of substitutable goods;
(b) the price of complementary goods;
(c) advertising expenditure.
2. If the line, P = −2/3Q + 6, is sketched with P on the horizontal axis and Q on the verti-
cal axis, find the gradient, m, and the vertical intercept, c.
3. If the demand function of a good is
2P + 3QD = 60
where P and QD denote price and quantity demanded, respectively, find the largest and
smallest values of P for which this function is economically meaningful.
4. The demand and supply functions of a good are given by
P = −5QD + 80
P = 2QS + 10
where P, QD and QS denote price, quantity demanded and quantity supplied, respectively.
(1) Find the equilibrium price and quantity
(a) graphically;
(b) algebraically.
(2) If the government deducts, as tax, 15% of the market price of each good, determine
the new equilibrium price and quantity.
Section 1.5 Supply and demand analysis 83
5. The supply and demand functions of a good are given by
P = QS + 8
P = −3QD + 80
where P, QS and QD denote price, quantity supplied and quantity demanded, respectively.
(a) Find the equilibrium price and quantity if the government imposes a fixed tax of
$36 on each good.
(b) Find the corresponding value of the government’s tax revenue.
6. The demand and supply functions for three interdependent commodities are
QD1 = 15 − P1 + 2P2 + P3
QD2 = 9 + P1 − P2 − P3
QD3 = 8 + 2P1 − P2 − 4P3
QS1 = −7 + P1
QS2 = −4 + 4P2
QS3 = −5 + 2P3
where QDi, QSi and Pi denote the quantity demanded, quantity supplied and price of good
i, respectively. Determine the equilibrium price and quantity for this three-commodity
model.
7. The demand and supply functions of a good are given by
P = −3QD + 60
P = 2QS + 40
respectively. If the government decides to impose a tax of $t per good, show that the
equilibrium quantity is given by
Q = 4 − 1/5 t
and write down a similar expression for the equilibrium price.
(a) If it is known that the equilibrium quantity is 3, work out the value of t. How much
of this tax is paid by the firm?
(b) If, instead of imposing a tax, the government provides a subsidy of $5 per good, find
the new equilibrium price and quantity.
8. The linear supply and demand functions for a good are given by
P = aQ + b and P = cQ + d
(a) State whether each of the values of the parameters a, b, c and d are positive or
negative.
(b) Find expressions, simplified as far as possible, for equilibrium price and quantity.
Section 1.6
Transposition of formulae
Objectives
At the end of this section you should be able to:
●
Manipulate formulae.
●Draw
●
a flow chart representing a formula.
Use a reverse flow chart to transpose a formula.
●Change
the subject of a formula involving several letters.
Mathematical modelling involves the use of formulae to represent the relationship between
economic variables. In microeconomics we have already seen how useful supply and demand
formulae are. These provide a precise relationship between price and quantity. For example,
the connection between price, P, and quantity, Q, might be modelled by
P = −4Q + 100
Given any value of Q it is trivial to deduce the corresponding value of P by merely replacing
the symbol Q by a number. A value of Q = 2, say, gives
P = −4 × 2 + 100
= −8 + 100
= 92
On the other hand, given P, it is necessary to solve an equation to deduce Q. For example,
when P = 40, the equation is
−4Q + 100 = 40
which can be solved as follows:
−4Q = −60 (subtract 100 from both sides)
Q = 15
(divide both sides by −4)
This approach is reasonable when only one or two values of P are given. However, if we are
given many values of P, it is clearly tedious and inefficient for us to solve the equation each
time to find Q. The preferred approach is to transpose the formula for P. In other words,
we rearrange the formula
P = an expression involving Q
into
Q = an expression involving P
Written this way round, the formula enables us to find Q by replacing P by a number. For the
specific formula
−4Q + 100 = P
Section 1.6 TrAnSPoSITIon oF ForMULAE
85
the steps are
24Q 5 P 2 100
Q5
P 2 100
24
(subtract 100 from both sides)
(divide both sides by 24)
Notice that
100
P 2 100
P
5
2
24
24 24
5 2 1 4 P 1 25
so the rearranged formula simplifies to
Q = −1/4 P + 25
If we now wish to find Q when P = 40, we immediately get
Q = −1/4 × 40 + 25
= −10 + 25
= 15
The important thing to notice about the algebra is that the individual steps are identical to
those used previously for solving the equation
−4Q + 100 = 40
i.e. the operations are again
‘subtract 100 from both sides’
followed by
‘divide both sides by −4’
practice problem
1. (a) Solve the equation
1
/2 Q + 13 = 17
State clearly exactly what operation you have performed to both sides at each stage
of your solution.
(b) By performing the same operations as part (a), rearrange the formula
1
/2 Q + 13 = P
into the form
Q = an expression involving P
(c) By substituting P = 17 into the formula derived in part (b), check that this agrees
with your answer to part (a).
86 Chapter 1 Linear Equations
In general, there are two issues concerning formula transposition. First, we need to decide
what to do to both sides of the given formula and the order in which they should be performed. Secondly, we need to carry out these steps accurately. The first of these is often the
more difficult. However, there is a logical strategy that can be used to help. To illustrate this,
consider the task of making Q the subject of
P = 1/3 Q + 5
that is, of rearranging this formula into the form
Q = an expression involving P
Imagine starting with a value of Q and using a calculator to work out P from
P = 1/3 Q + 5
The diagram that follows shows that two operations are required and indicates the order in
which they must be done. This diagram is called a flow chart.
Q
P
divide by 3
add 5
To go backwards from P to Q we need to undo these operations. Now the reverse of ‘divide
by 3’ is ‘multiply by 3’ and the reverse of ‘add 5’ is ‘subtract 5’, so the operations needed to
transpose the formula are as follows:
Q
P
multiply by 3
subtract 5
This diagram is called a reverse flow chart. The process is similar to that of unwrapping
a parcel (or peeling an onion); you start by unwrapping the outer layer first and work
inwards. If we now actually perform these steps in the order specified by the reverse flow
chart, we get
/3 Q + 5 = P
1
1
/3 Q
Q
=P−5
(subtract 5 from both sides)
= 3(P − 5)
(multiply both sides by 3)
The rearranged formula can be simplified by multiplying out the brackets to give
Q = 3P − 15
Incidentally, if you prefer, you can actually use the reverse flow chart itself to perform the
algebra for you. All you have to do is to pass the letter P through the reverse flow chart.
Working from right to left gives
Q 5 3(P ] 5)
P]5
multiply by 3
P
subtract 5
Notice that taking P as the input to the box ‘subtract 5’ gives the output P − 5, and if the
whole of this is taken as the input to the box ‘multiply by 3’, the final output is the answer,
3(P − 5). Hence
Q = 3(P − 5)
Section 1.6 Transposition of formulae 87
Example
Make x the subject of
x
4
(b) y 5
5
2x 1 1
(a) y 5
Solution
(a) To go from x to y, the operations are
x
y
divide by 5
square root
so the steps needed to transpose the formula are
x
y
multiply by 5
square
The algebraic details are as follows:
x
5y
5
x
5 y2
5
x 5 5y2
(square both sides)
(multiply both sides by 5)
Hence the transposed formula is
x = 5y2
Alternatively, if you prefer, the reverse flow chart can be used directly to obtain
x 5 5y 2
y2
multiply by 5
y
square
Hence
x = 5y2
(b) The forwards flow chart is
x
y
multiply by 2
add 1
reciprocate
multiply by 4
so the reverse flow chart is
x
y
divide by 2
subtract 1
reciprocate
divide by 4
➜
88 Chapter 1 Linear Equations
The algebraic details are as follows:
4
5y
2x 1 1
1
y
5
2x 1 1 4
(divide both sides by 4)
4
2x 1 1 5 y
(reciprocate both sides)
4
2x 5 y 2 1
5
1 4
21
2 y
(subtract 1 from both sides)
(divide both sides by 2)
which can be simplified, by multiplying out the brackets, to give
x5
2 1
2
2
y
Again, the reverse flow chart can be used directly to obtain
x5
1 4
( ] 1)
2 y
4
]1
y
divide by 2
y
4
4
y
subtract 1
reciprocate
y
divide by 4
Practice Problem
2. Use flow charts to make x the subject of the following formulae:
(a) y = 6x2 (b) y 5
1
7x 2 1
The following example contains two difficult instances of transposition. In both cases the
letter x appears more than once on the right-hand side. If this happens, the technique based
on flow charts cannot be used. However, it may still be possible to perform the manipulation
even if some of the steps may not be immediately obvious.
Example
Transpose the following formulae to express x in terms of y:
(a) ax = bx + cy + d (b) y 5
Solution
(a) In the formula
ax = bx + cy + d
x11
x22
Section 1.6 Transposition of formulae 89
there are terms involving x on both sides, and since we are hoping to rearrange this into
the form
x = an expression involving y
it makes sense to collect the xs on the left-hand side. To do this we subtract bx from
both sides to get
ax − bx = cy + d
Notice that x is a common factor of the left-hand side, so the distributive law can be
applied ‘in reverse’ to take the x outside the brackets: that is,
(a – b)x = cy + d
Finally, both sides are divided by a − b to get
x5
cy 1 d
a2b
which is of the desired form.
(b) It is difficult to see where to begin with the formula
y5
x11
x22
because there is an x in both the numerator and the denominator. Indeed, the thing that
is preventing us getting started is precisely the fact that the expression is a fraction. We
can, however, remove the fraction simply by multiplying both sides by the denominator
to get
(x − 2)y = x + 1
and if we multiply out the brackets, then
xy − 2y = x + 1
We want to rearrange this into the form
x = an expression involving y
so we collect the xs on the left-hand side and put everything else on to the right-hand
side. To do this we first add 2y to both sides to get
xy = x + 1 + 2y
and then subtract x from both sides to get
xy − x = 1 + 2y
The distributive law can now be applied ‘in reverse’ to take out the common factor of
x: that is,
(y − 1)x = 1 + 2y
Finally, dividing through by y − 1 gives
x5
1 1 2y
y21
90 Chapter 1 Linear Equations
Advice
This example contains some of the hardest algebraic manipulation seen so far in this text. I hope that
you managed to follow the individual steps. However, it all might appear as if we have ‘pulled rabbits
out of hats’. You may feel that, if left on your own, you are never going to be able to decide what to
do at each stage. Unfortunately, there is no watertight strategy that always works, although the following five-point plan is worth considering if you get stuck.
To transpose a given formula of the form
y = an expression involving x
into a formula of the form
x = an expression involving y
you proceed as follows:
Step 1 Remove fractions.
Step 2
Multiply out the brackets.
Step 3
Collect all of the xs on to the left-hand side.
Step 4
Take out a factor of x.
Step 5
Divide by the coefficient of x.
You might find it helpful to look back at the previous example in the light of this strategy. In part (b)
it is easy to identify each of the five steps. Part (a) also used this strategy, starting with the third step.
Example
Make x the subject of
ax 1 b
cx 1 d
y5
Solution
In this formula there is a square root symbol surrounding the right-hand side. This can be
removed by squaring both sides to get
y2 5
ax 1 b
cx 1 d
We now apply the five-step strategy:
Step 1 (cx + d)y2 = ax + b
Step 2 cxy2 + dy2 = ax + b
Step 3 cxy2 − ax = b − dy2
Step 4 (cy2 − a)x = b − dy2
Step 5
x5
b 2 dy2
cy2 2 a
Section 1.6 Transposition of formulae 91
Practice Problem
3. Transpose the following formulae to express x in terms of y:
(a) x − ay = cx + y
(b) y 5
x22
x14
Key Terms
Flow chart A diagram consisting of boxes of instructions indicating a sequence of operations and
their order.
Reverse flow chart A flow chart indicating the inverse of the original sequence of operations in
reverse order.
Transpose a formula The rearrangement of a formula to make one of the other letters the subject.
Exercise 1.6
1. Make Q the subject of
P = 2Q + 8
Hence find the value of Q when P = 52.
2. Write down the formula representing each of the following flow charts
x
y
(a)
double
add 5
x
y
(b)
add 5
double
x
y
(c)
square
reciprocate
multiply by 5
x
y
(d)
add 4
square
multiply by 2
subtract 3
3. Draw flow charts for each of the following formulae:
(a) y = 5x + 3
(e) y 5
x
17
2
(b) y = 5(x + 3)
(f) y 5
2
x
(c) y = 6x – 9
(g) y 5
(d) y = 4x2 – 6
1
x13
4. Make x the subject of each of the following formulae:
(a) y = 9x – 6
(d) y 5
x
18
5
(b) y = (x + 4)/3
(e) y 5
1
x12
(c) y 5
x
2
(f) y 5
4
3x 2 7
➜
92 Chapter 1 Linear Equations
5. Transpose the formulae:
(a) Q = aP + b to express P in terms of Q
(b) Y = aY + b + I to express Y in terms of I
(c) Q 5
1
to express P in terms of Q
aP 1 b
6. Make x the subject of the formula
y5
3
22
x
7. In business, the economic order quantity is Q 5
2DR
.
H
(a) Make D the subject of this formula.
(b) Make H the subject of this formula.
Exercise 1.6*
1. (1) Draw flow charts for each of the following formulae:
(a) y = 9x + 1
(b) y = 3 – x
4
(e) y 5 2
x 18
(d) y = ∙3x + 5
(c) y = 5x2 – 8
(2) Hence, or otherwise, express x in terms of y in each case.
2. Make x the subject of the following formulae:
(a)
a
c
1b5
x
x
(d) a
(b) a 2 x 5
x2n
a2
5
m
b
(e)
b1x
a
x2m 1
5
n
m
(c) e 1 x 1 f 5 g
(f)
x1a
x2b
5
3. Transpose the formula
V5
5t 1 1
t21
to express t in terms of V.
Hence, or otherwise, find the value of t when V = 5.6.
4. Make r the subject of the formula
r
S 5 P 11
100
n
5. Rearrange the formula
Y5
2aT 1 b 1 I 1 G
1 2 a 1 at
to make each of the following letters the subject:
(a) G (b) T (c) t (d) a
b
a
Section 1.7
National income determination
Objectives
At the end of this section you should be able to:
●
Identify and sketch linear consumption functions.
●
Identify and sketch linear savings functions.
●
Set up simple macroeconomic models.
●Calculate
●
equilibrium national income.
Analyse IS and LM schedules.
Macroeconomics is concerned with the analysis of economic theory and policy at a national
level. In this section we focus on one particular aspect known as national income deter­
mination. We describe how to set up simple models of the national economy which enable
equilibrium levels of income to be calculated. Initially we assume that the economy is divided
into two sectors, households and firms. Firms use resources such as land, capital, labour and raw
materials to produce goods and services. These resources are known as factors of production and
are taken to belong to households. National income represents the flow of income from firms
to households given as payment for these factors. Households can then spend this money in
one of two ways. Income can be used for the consumption of goods produced by firms, or it
can be put into savings. Consumption, C, and savings, S, are therefore functions of income, Y:
that is,
C = f(Y)
S = g(Y)
for some appropriate consumption function, f, and savings function, g. Moreover, C and S are
normally expected to increase as income rises, so f and g are both increasing functions.
We begin by analysing the consumption function. As usual we need to quantify the precise
relationship between C and Y. If this relationship is linear, then a graph of a typical consumption
function is shown in Figure 1.22. It is clear from this graph that if
C = aY + b
then a > 0 and b > 0. The intercept b is the level of consumption when there is no income
(that is, when Y = 0) and is known as autonomous consumption. The slope, a, is the change in
C brought about by a 1-unit increase in Y and is known as the marginal propensity to consume
(MPC). As previously noted, income is used up in consumption and savings, so that
Y=C+S
It follows that only a proportion of the 1-unit increase in income is consumed; the rest goes into
savings. Hence the slope, a, is generally smaller than 1: that is, a < 1. It is standard practice in
mathematics to collapse the two separate inequalities a > 0 and a < 1 into the single inequality
0<a<1
94 Chapter 1 Linear Equations
C
C 5 aY 1 b
b
Y
Figure 1.22
The relation
Y=C+S
enables the precise form of the savings function to be determined from any given consumption
function.
To be specific, suppose that the consumption function is given by
C = 0.6Y + 10
The graph is sketched in Figure 1.23 using the fact that it passes through (0, 10) and (40, 34).
To find the savings function we use the relation
Y=C+S
which gives
(subtract C from both sides)
S = Y − C.
= Y − (0.6Y + 10)
(substitute C)
= Y − 0.6Y − 10
(multiply out the brackets)
= 0.4Y − 10
(collect terms)
C
(40, 34)
C 5 0.6Y 1 10
10
Y
Figure 1.23
Section 1.7 National income determination 95
S
S 5 0.4Y ] 10
25
Y
]10
Figure 1.24
The savings function is also linear. Its graph has intercept −10 and slope 0.4. This is sketched
in Figure 1.24 using the fact that it passes through (0, −10) and (25, 0).
Practice Problem
1. Determine the savings function that corresponds to the consumption function
C = 0.8Y + 25
For the general consumption function
C = aY + b
we have
S=Y−C
= Y − (aY + b)
(substitute C)
= Y − aY − b
(multiply out the brackets)
= (1 − a)Y − b
(take out a common factor of Y )
The slope of the savings function is called the marginal propensity to save (MPS) and is given
by 1 − a: that is,
MPS = 1 − a = 1 − MPC
Moreover, since a < 1, we see that the slope, 1 − a, is positive. Figure 1.25 shows the graph
of this savings function. One interesting feature, which contrasts with other economic functions considered so far, is that it is allowed to take negative values. In particular, note that
autonomous savings (that is, the value of S when Y = 0) are equal to −b, which is negative
because b > 0. This is to be expected because whenever consumption exceeds income, households must finance the excess expenditure by withdrawing savings.
Advice
The result, MPC + MPS = 1, is always true, even if the consumption function is non-linear. A proof
of this generalisation can be found in Section 4.3.3 in Chapter 4.
96 Chapter 1 Linear Equations
S
S 5 (1 ] a)Y ] b
Y
]b
Figure 1.25
The simplest model of the national economy is illustrated in Figure 1.26, which shows the
circular flow of income and expenditure. This is fairly crude, since it fails to take into account
government activity or foreign trade. In this diagram, investment, I, is an injection into the
circular flow in the form of spending on capital goods.
Let us examine this more closely and represent the diagrammatic information in symbols.
Consider first the box labelled ‘Households’. The flow of money entering this box is Y and
the flow leaving it is C + S. Hence we have the familiar relation
Y=C+S
Injections:
investment
l
C
Firms
Y
Expenditure:
consumption of
domestically
produced goods
Income; payments
for factors of
production
Households
C
Y
S
Withdrawals:
savings
Figure 1.26
Section 1.7 National income determination 97
For the box labelled ‘Firms’ the flow entering it is C + I and the flow leaving it is Y, so
Y=C+I
Suppose that the level of investment that firms plan to inject into the economy is known to
be some fixed value, I*. If the economy is in equilibrium, the flow of income and expenditure
balance so that
Y = C + I*
From the assumption that the consumption function is
C = aY + b
for given values of a and b, these two equations represent a pair of simultaneous equations for
the two unknowns Y and C. In these circumstances C and Y can be regarded as endogenous
variables, since their precise values are determined within the model, whereas I* is fixed outside the model and is exogenous.
Example
Find the equilibrium level of income and consumption if the consumption function is
C = 0.6Y + 10
and planned investment I = 12.
Solution
We know that
Y=C+I
(from theory)
C = 0.6Y + 10
(given in problem)
(given in problem)
I = 12
If the value of I is substituted into the first equation, then
Y = C + 12
The expression for C can also be substituted to give
Y = 0.6Y + 10 + 12
Y = 0.6Y + 22
0.4Y = 22
Y = 55
(subtract 0.6Y from both sides)
(divide both sides by 0.4)
The corresponding value of C can be deduced by putting this level of income into the
consumption function to get
C = 0.6(55) + 10 = 43
The equilibrium income can also be found graphically by plotting expenditure against
income. In this example the aggregate expenditure, C + I, is given by 0.6Y + 22. This is
sketched in Figure 1.27 using the fact that it passes through (0, 22) and (80, 70). Also
sketched is the ‘45° line’, so called because it makes an angle of 45° with the horizontal.
This line passes through the points (0, 0), (1, 1), . . . , (50, 50) and so on. In other words,
➜
98 Chapter 1 Linear Equations
at any point on this line expenditure and income are in balance. The equilibrium income
can therefore be found by inspecting the point of intersection of this line and the aggregate
expenditure line, C + I. From Figure 1.27 this occurs when Y = 55, which is in agreement
with the calculated value.
Expenditure
(80, 70)
55
C 1 l line
22
45° line
55
Income
Figure 1.27
Practice Problem
2. Find the equilibrium level of income if the consumption function is
C = 0.8Y + 25
and planned investment I = 17. Calculate the new equilibrium income if planned
investment rises by 1 unit.
To make the model more realistic, let us now include government expenditure, G, and taxation,
T, in the model. The injections box in Figure 1.26 now includes government expenditure in
addition to investment, so
Y=C+I+G
We assume that planned government expenditure and planned investment are autonomous
with fixed values G* and I*, respectively, so that in equilibrium
Y = C + I* + G*
The withdrawals box in Figure 1.26 now includes taxation. This means that the income that
households have to spend on consumer goods is no longer Y but rather Y − T (income less
tax), which is called disposable income, Yd. Hence
C = aYd + b
Section 1.7 National income determination 99
with
Yd = Y − T
In practice, the tax will either be autonomous (T = T* for some lump sum T*) or be a proportion
of national income (T = tY for some proportion t), or a combination of both (T = tY + T*).
Example
Given that
G = 20
I = 35
C = 0.9Yd + 70
T = 0.2Y + 25
calculate the equilibrium level of national income.
Solution
At first sight this problem looks rather forbidding, particularly since there are so many
variables. However, all we have to do is to write down the relevant equations and to substitute systematically one equation into another until only Y is left.
We know that
Y=C+I+G
(from theory)
(1)
G = 20
(given in problem)
(2)
I = 35
(given in problem)
(3)
C = 0.9Yd + 70
(given in problem)
(4)
T = 0.2Y + 25
(given in problem)
(5)
Yd = Y − T
(from theory)
(6)
This represents a system of six equations in six unknowns. The obvious thing to do is to
put the fixed values of G and I into equation (1) to get
Y = C + 35 + 20 = C + 55
(7)
This has at least removed G and I, so there are only three more variables (C, Yd and T)
left to eliminate. We can remove T by substituting equation (5) into (6) to get
Yd = Y − (0.2Y + 25)
= Y − 0.2Y − 25
= 0.8Y − 25
(8)
and then remove Yd by substituting equation (8) into (4) to get
C = 0.9(0.8Y − 25) + 70
= 0.72Y − 22.5 + 70
= 0.72Y + 47.5
(9)
➜
100 Chapter 1 Linear Equations
We can eliminate C by substituting equation (9) into (7) to get
Y = C + 55
= 0.72Y + 47.5 + 55
= 0.72Y + 102.5
Finally, solving for Y gives
0.28Y = 102.5 (subtract 0.72Y from both sides)
Y = 366
(divide both sides by 0.28)
Practice Problem
3. Given that
G = 40
I = 55
C = 0.8Yd + 25
T = 0.1Y + 10
calculate the equilibrium level of national income.
To conclude this section we return to the simple two-sector model:
Y =C+I
C = aY + b
Previously, the investment, I, was taken to be constant. It is more realistic to assume that
planned investment depends on the rate of interest, r. As the interest rate rises, so investment
falls and we have a relationship
I = cr + d
where c < 0 and d > 0. Unfortunately, this model consists of three equations in the four
unknowns Y, C, I and r, so we cannot expect it to determine national income uniquely. The
best we can do is to eliminate C and I, say, and to set up an equation relating Y and r. This
is most easily understood by an example. Suppose that
C = 0.8Y + 100
I = −20r + 1000
We know that the commodity market is in equilibrium when
Y=C+I
Substitution of the given expressions for C and I into this equation gives
Y = (0.8Y + 100) + (−20r + 1000)
= 0.8Y − 20r + 1100
Section 1.7 National income determination 101
which rearranges as
0.2Y + 20r = 1100
This equation, relating national income, Y, and interest rate, r, is called the IS schedule.
We obviously need some additional information before we can pin down the values of Y
and r. This can be done by investigating the equilibrium of the money market. The money
market is said to be in equilibrium when the supply of money, MS, matches the demand for
money, MD: that is, when
MS = MD
There are many ways of measuring the money supply. In simple terms it can be thought
of as consisting of the notes and coins in circulation, together with money held in bank
deposits. The level of MS is assumed to be controlled by the central bank and is taken to be
autonomous, so that
MS = M*S
for some fixed value M *S .
The demand for money comes from three sources: transactions, precautions and speculations.
The transactions demand is used for the daily exchange of goods and services, whereas the
precautionary demand is used to fund any emergencies requiring unforeseen expenditure. Both
are assumed to be proportional to national income. Consequently, we lump these together
and write
L1 = k1Y
where L1 denotes the aggregate transaction–precautionary demand and k1 is a positive constant. The speculative demand for money is used as a reserve fund in case individuals or firms
decide to invest in alternative assets such as government bonds. In Chapter 3 we show that,
as interest rates rise, speculative demand falls. We model this by writing
L2 = k2r + k3
where L2 denotes speculative demand, k2 is a negative constant and k3 is a positive constant.
The total demand, MD, is the sum of the transaction–precautionary demand and speculative
demand: that is,
MD = L1 + L2
= k1Y + k2 r + k3
If the money market is in equilibrium, then
MS = MD
that is,
M*S = k1Y + k2r + k3
This equation, relating national income, Y, and interest rate, r, is called the LM schedule. If
we assume that equilibrium exists in both the commodity and money markets, then the IS
and LM schedules provide a system of two equations in two unknowns, Y and r. These can
easily be solved either by elimination or by graphical methods.
102 Chapter 1 Linear Equations
Example
Determine the equilibrium income and interest rate given the following information about
the commodity market:
C = 0.8Y + 100
I = −20r + 1000
and the money market:
MS = 2375
L1 = 0.1Y
L2 = −25r + 2000
What effect would a decrease in the money supply have on the equilibrium levels of Y and r?
Solution
The IS schedule for these particular consumption and investment functions has already been
obtained in the preceding text. It was shown that the commodity market is in equilibrium
when
0.2Y + 20r = 1100
(1)
For the money market we see that the money supply is
MS = 2375
and that the total demand for money (that is, the sum of the transaction–precautionary
demand, L1, and the speculative demand, L2) is
MD = L1 + L2 = 0.1Y − 25r + 2000
The money market is in equilibrium when
MS = MD
that is,
2375 = 0.1Y − 25r + 2000
The LM schedule is therefore given by
0.1Y − 25r = 375
(2)
Equations (1) and (2) constitute a system of two equations for the two unknowns Y and r.
The steps described in Section 1.4 can be used to solve this system:
Step 1
Double equation (2) and subtract from equation (1) to get
0.2Y 1 20r 5 1100
0.2Y 2 50r 5 750 2
70r 5 350
Step 2
Divide both sides of equation (3) by 70 to get
r=5
(3)
Section 1.7 National income determination 103
Step 3
Substitute r = 5 into equation (1) to get
0.2Y + 100 = 1100
(subtract 100 from both sides)
0.2Y = 1000
(divide both sides by 0.2)
Y = 5000
Step 4
As a check, equation (2) gives
0.1(5000) − 25(5) = 375 ✓
The equilibrium levels of Y and r are therefore 5000 and 5, respectively.
To investigate what happens to Y and r as the money supply falls, we could just take a
smaller value of MS such as 2300 and repeat the calculations. However, it is more instructive
to perform the investigation graphically. Figure 1.28 shows the IS and LM curves plotted
on the same diagram with r on the horizontal axis and Y on the vertical axis. These lines
intersect at (5, 5000), confirming the equilibrium levels of interest rate and income obtained
by calculation. Any change in the money supply will obviously have no effect on the IS
curve. On the other hand, a change in the money supply does affect the LM curve. To see
this, let us return to the general LM schedule
k1Y + k2r + k3 = M *S
Y
LM
curve
5500
5000
3750
IS curve
5
Figure 1.28
55
r
➜
104 Chapter 1 Linear Equations
and transpose it to express Y in terms of r:
k1Y = −k2r − k3 + M *S
Y5
2k2
2k3 1 M *S
r1
k1
k1
(subtract k2r + k3 from both sides)
(divide both sides by k1)
Expressed in this form, we see that the LM schedule has slope −k2 /k1 and intercept
(−k3 + M *S )/k1.
Any decrease in M *S therefore decreases the intercept (but not the slope), and the LM
curve shifts downwards. This is indicated by the dashed line in Figure 1.28. The point of
intersection shifts both downwards and to the right. We deduce that, as the money supply
falls, interest rates rise and national income decreases (assuming that both the commodity
and money markets remain in equilibrium).
Advice
It is possible to produce general formulae for the equilibrium level of income in terms of various
parameters used to specify the model. As you might expect, the algebra is a little harder, but it does
allow for a more general investigation into the effects of varying these parameters. We will return to
this in Section 5.3.
Practice Problem
4. Determine the equilibrium income, Y, and interest rate, r, given the following informa-
tion about the commodity market
C = 0.7Y + 85
I = −50r + 1200
and the money market
MS = 300
L1 = 0.2Y
L2 = −40r + 30
Sketch the IS and LM curves on the same diagram. What effect would an increase in
the value of autonomous investment have on the equilibrium values of Y and r?
Section 1.7 National income determination 105
Key Terms
Autonomous consumption The level of consumption when there is no income.
Autonomous savings The withdrawals from savings when there is no income.
Consumption function The relationship between national income and consumption.
Disposable income Household income after the deduction of taxes and the addition of benefits.
Factors of production
and raw materials.
The inputs to the production of goods and services: land, capital, labour
Government expenditure The total amount of money spent by government on defence, education,
health, police, etc.
Investment
The creation of output not for immediate consumption.
IS schedule The equation relating national income and interest rate based on the assumption of
equilibrium in the goods market.
LM schedule The equation relating national income and interest rate based on the assumption of
equilibrium in the money market.
Marginal propensity to consume The fraction of a rise in national income which goes into consumption. It is the slope of the consumption function.
Marginal propensity to save The fraction of a rise in national income which goes into savings. It
is the slope of the savings function.
Money supply
The notes and coins in circulation together with money held in bank deposits.
National income The flow of money from firms to households.
Precautionary demand for money
seen future expenditure.
Money held in reserve by individuals or firms to fund unfore-
Speculative demand for money Money held back by firms or individuals for the purpose of
investing in alternative assets, such as government bonds, at some future date.
Taxation Money paid to government based on an individual’s income and wealth (direct taxation)
together with money paid by suppliers of goods or services based on expenditure (indirect taxation).
Transactions demand for money
Money used for everyday transactions of goods and services.
Exercise 1.7
1. If the consumption function is given by C = 4200 + 0.75Y, state the marginal propen-
sity to consume and deduce the marginal propensity to save.
2. If the national income, Y, is 1000 units, then consumption, C, is 800 units. Also when-
ever income rises by 100, consumption increases by 70. Assuming that the consumption
function is linear:
(a) state the marginal propensity to consume and deduce the marginal propensity to
save;
(b) find an expression for C in terms of Y.
3. If the consumption function is given by
C = 0.7Y + 40
state the values of
➜
106 Chapter 1 Linear Equations
(a) autonomous consumption;
(b) marginal propensity to consume.
Transpose this formula to express Y in terms of C and hence find the value of Y when
C = 110.
4. Write down expressions for the savings function given that the consumption function is
(a) C = 0.9Y + 72
(b) C = 0.8Y + 100
5. For a closed economy with no government intervention, the consumption function is
C = 0.6Y + 30
and planned investment is
I = 100
Calculate the equilibrium level of
(a) national income;
(b) consumption;
(c) savings.
6. A consumption function is given by C = aY + b.
It is known that when Y = 10, the value of C is 28, and that when Y = 30, the value
of C is 44.
By solving a pair of simultaneous equations, find the values of a and b, and deduce
that the corresponding savings function is given by
S = 0.2Y − 20
Determine the equilibrium level of income when planned investment I = 13.
7. Given that
G = 50
I = 40
C = 0.75Yd + 45
T = 0.2Y + 80
calculate the equilibrium level of national income.
Exercise 1.7*
1. Write down an expression for the savings function, simplified as far as possible, given
that the consumption function is
(a) C = 0.7Y + 30
(b) C 5
Y 2 1 500
Y 1 10
Section 1.7 National income determination 107
2. If
C = aY + b
Y=C+I
I = I*
show that
Y5
b 1 I*
12a
and obtain a similar expression for C in terms of a, b and I*.
3. Transpose the formula
Y5
b 1 I*
12a
to express a in terms of Y, b and I*.
4. An open economy is in equilibrium when
Y=C+I+G+X−M
where
Y = national income
C = consumption
I = investment
G = government expenditure
X = exports
M = imports
Determine the equilibrium level of income given that
C = 0.8Y + 80
I = 70
G = 130
X = 100
M = 0.2Y + 50
5. Given that
consumption, C = 0.8Y + 60
investment, I = −30r + 740
money supply,
MS = 4000
transaction–precautionary demand for money, L1 = 0.15Y
speculative demand for money, L2 = −20r + 3825
determine the values of national income, Y, and interest rate, r, on the assumption that
both the commodity and the money markets are in equilibrium.
➜
108 Chapter 1 Linear Equations
6. Consider the national income model
Y=C+I
C = aYd + 50
I = 24
Yd = Y − T
T = 20
Show that the equilibrium level of national income is given by
Y5
74 2 20a
12a
Transpose this equation to express a in terms of Y.
Hence, or otherwise, find the value of a for which Y = 155 and find the value of C.
7. Consider the national income model
Y = C + I* + G*
C = a(Y − T),
0<a<1
T = tY,
0<t<1
Show that
Y5
I* 1 G*
1 1 a(t 2 1)
and hence state what happens to Y when
(a) G* increases (b) t increases
Formal mathematics
The approach adopted in this text is very informal. Emphasis is placed throughout on making
both the mathematics and economic applications as accessible as possible. Hopefully this will
help you to understand and enjoy the subject. However, it could well be that your lecturers
use more formal language and notation. For this reason we conclude each chapter with a brief
discussion of more rigorous mathematical language and ideas which provide the framework
for more advanced courses which you may take in future years.
In Section 1.2 a number line was used to represent the solutions of linear inequalities.
The set of all numbers which lie between two values on the number line is called an interval
and the notation that we use is summarised in the following table:
Interval
Notation
a≤x≤b
[a, b]
a<x<b
(a, b)
a≤x<b
[a, b)
a<x≤b
(a, b]
The first interval, which includes both end points, is called a closed interval and the second,
which excludes the end points, is called an open interval. Using this notation we would write
[2, 5] as an abbreviation for 2 ≤ x ≤ 5 and (3, 9] as an abbreviation for 3 < x ≤ 9. It is also
possible to use this notation to include intervals which are unbounded. For example, the interval x ≥ 6, which consists of all numbers to the right of 6 on the number line (including
6 itself), can be written as [6, ∞). The symbol ∞ (infinity) is not a number but merely indicates
that the interval goes on forever without an upper limit. In this notation we could write the
complete set of numbers as (−∞, ∞).
It is often convenient in mathematics to ignore the sign of a number and deliberately make
it positive. This is called the absolute value or modulus of a number, x, and is written u x u. In
this notation we have
u−5 u = 5 and u 4 u = 4
If a number is negative, we change the sign to make it positive, whereas if the number is
already positive, it is unchanged, so we have
u x u = −x if x < 0
and u x u = x if x ≥ 0
Example
Solve the following inequalities giving your answer using interval notation.
(a) u x u ≤ 6 (b) u 2x − 1 u < 3
➜
110 Chapter 1 Linear Equations
Solution
(a) If the absolute value of a number x is less than or equal to 6, then the number itself
must be between −6 and 6 inclusive, so the solution of the inequality u x u ≤ 6 can be
expressed in interval notation as [−6, 6].
(b) If the absolute value of a number 2x − 1 is less than 3, then the number itself must be
between −3 and 3, so we need to solve
−3 < 2x − 1 < 3
−2 < 2x < 4
(add 1 to both sides)
−1 < x < 2
(divide both sides by 2)
In interval notation this is written (−1, 2).
In Section 1.5 we introduced the key concept of a function. The domain of a function
consists of the set of all possible inputs, and the set of corresponding outputs is called the
range of the function. Sometimes we are forced to restrict the inputs to a function because
it is impossible to evaluate the function for certain values of x. For example, it is impossible
to evaluate f (4) for the function
f (x) 5
1
x24
If you try to substitute x = 4 into this function, you get ‘1/0’ which is meaningless. On a
calculator you get a message ‘division by zero’ error. For this function the domain consists of
all numbers, x Þ 4. It is impossible to evaluate f(x) 5 x 2 3 for values of x which are less than
3 because it is impossible to find the square root of negative numbers. For this function the
domain consists of all numbers, x ≥ 3, so in interval notation, the domain is [3, ∞).
Sometimes we choose to restrict the domain of a function so that it makes economic sense.
For example, in the demand function, P = −2Q + 50, sketched in Figure 1.15, we would
choose the domain as [0, 25]. We obviously need to exclude negative quantities from the
domain and the graph shows that if Q exceeds 25, then P would be negative which is nonsense. From a purely mathematical point of view, there is no problem inputting any number
into the function. The line drawn in Figure 1.15 just goes on forever in each direction.
However, for the function to be economically meaningful we would choose to restrict the
domain to [0, 25]. If we do this, then the vertical axis in Figure 1.15 also shows that the corresponding outputs consist of all possible numbers between 0 and 50. In other words, the
range of the function is [0, 50].
Mathematical notation exists to make the subject precise and unambiguous. It also provides
a convenient shorthand to reduce the amount of writing that we need to do. The following
table lists symbols which are often used when presenting mathematical arguments, and you
may encounter these when reading higher-level books and journals.
Symbol
Meaning
[
therefore
∃
there exists
∀
Q
for all
3
is equivalent to
implies
Formal mathematics 111
The implies symbol provides a concise way of linking mathematical statements together. We
could write the following when solving equations:
2x − 8 = 10 Q
2x = 18
Q
x=9
Use of this symbol has allowed us to fit a string of mathematical steps on a single line of working.
The equivalence symbol 3 can be used whenever the implication works both ways. Technically,
you could replace Q by 3 in the algebra above since the steps are reversible. However, it would
be pointless to do so, since this piece of mathematics is clearly designed to solve an equation
which you do by working from left to right. However, not all implications are reversible, so
you must be careful when using the double implication sign. For example, the line
x = −5
Q
uxu = 5
is true but the reverse is not necessarily true since we have the alternative option of x = 5, so
it would not be correct to use 3 here. On the other hand, use of 3 in the following line is
correct and conveys the important message that two numbers multiply to zero if and only if
at least one of them is zero.
xy = 0 3
x = 0 or y = 0.
Key Terms
Absolute value
The positive value or magnitude of a number.
Closed interval
The set of all real numbers between and including two given numbers: a ≤ x ≤ b.
Domain
The numbers which are used as inputs to a function.
Interval The set of all real numbers between (and possibly including) two given numbers.
Modulus
The positive value or magnitude of a number.
Open interval The set of all real numbers between but excluding two given numbers: a < x < b.
Range
The numbers which form the set of outputs from a function.
Multiple choice questions
Question 1
A graph of the demand function, 3P + 4Q = 9 is plotted with Q on the horizontal axis.
Find the slope of the line and the intercept on the P-axis.
A. Slope is 9/4 and intercept is −3
B. Slope is 3/4 and intercept is 9/4
C. Slope is 4/3 and intercept is 1/3
D. Slope is −4/9 and intercept is 1/3
E. Slope is −4/3 and intercept is 3
Question 2
Multiply out the brackets and collect like terms in the expression 5(3x − 4y) − (2x − 3y).
A. 13x − 23y
B. 13x − 17y
C. 13x − y
D. 13x − 7y
E. 13x − 20y
Question 3
If the supply function is Q = 4P − 25, determine the lowest possible price that could be
charged.
A. 25
B. 0
C. 25/4
D. 4/25
E. 100
D. x ≤ 1/3
E. x ≥ −3
Question 4
Solve the inequality: 2(x + 3) − (5x + 1) ≥ 4
A. x ≥ −1/3
B. x ≤ −1
Question 5
Solve the simultaneous equations:
4x − 2y = 3
6x − 3y = 4
A. No solution
B. Infinitely many solutions
C. x = 0, y = −1.5
D. x = 1, y = 0.5
E. x = −0.5, y = −2.5
C. x ≥ 1
Multiple choice questions 113
Question 6
The demand and supply functions of a good are given by:
P = −3QD + 75
P = 2QS + 15
Determine the equilibrium price.
A. 39
B. 12
C. 60
D. 15
E. 27
Question 7
If the government imposes a fixed tax of $5 on each good sold in Question 6, determine
the post-tax equilibrium price.
A. 40
B. 13
C. 41
D. 11
E. 42
Question 8
A demand function is given by P = aQ + b. Work out the missing number in the table.
Q
10
P
6
A. 5.0
14
25
4.5
B. 5.6
C. 4.8
D. 5.5
E. 3.6
Question 9
Consider the macroeconomic model:
G
I
C
Y
= 30
= 90
= 0.8Y + 20
=C+G+I
(government expenditure)
(planned investment)
(consumption)
(equilibrium)
Work out the change in the value of national income Y when government expenditure
rises by 1 unit.
A. 13
B. 10
C. 8
D. 5
E. 18
Question 10
Find the value of national income Y when the consumption function is C = 0.75Y + 4 and
savings are 3.5.
A. 30
B. 2
C. 7
D. 44
E. 54
114 Chapter 1 Linear Equations
Question 11
If the supply and demand functions are given by
P = aQS + b
P = cQD + d
what conditions must be imposed on the parameters, a, b, c and d for these equations to
be economically meaningful and for the equilibrium to exist?
A. a > 0, b > 0, c < 0, d < 0, d < b
B. a < 0, b < 0, c > 0, d > 0, d < b
C. a < 0, b > 0, c > 0, d < 0, d < b
D. a > 0, b > 0, c < 0, d > 0, d > b
E. a > 0, b < 0, c > 0, d > 0, d > b
Question 12
Transpose the equation x 5
A. y 5
3 2 7x
x25
B. y 5
5y 1 12
to express y in terms of x.
y13
9
x25
C. y 5
12
x25
D. y 5
3(4 2 x)
x25
E. y 5
3x 1 8
x25
Question 13
Express the following as a single fraction:
A.
4
1
2
x12 x21
3x 1 1
3(x 2 2)
5x 2 1
5x 2 2
5x 1 7
B.
C.
D.
E.
(x 2 1)(x 1 2)
(x
2
1)(x
1
2)
(x
2
1)(x 1 2)
(x 2 1)(x 1 2)
(x 2 1)(x 1 2)
Question 14
Solve the simultaneous equations
ax + by = c
dx + ey = f
A. x 5
bf 2 ce
cd 2 bf
, y5
ae 2 bd
ae 2 bd
B. x 5
ce 1 bf
af 2 cd
, y5
bd 2 ae
bd 2 ae
C. x 5
ce 2 bf
cd 2 af
, y5
bd 2 ae
bd 2 ae
D. x 5
bf 2 ce
af 2 cd
, y5
bd 2 ae
bd 2 ae
E. x 5
ce 2 bf
af 2 cd
, y5
ae 2 bd
ae 2 bd
Multiple choice questions 115
Question 15
If the following system of equations has infinitely many solutions, find the value of k.
x − 2y + 5z = 7
2x + 3y − 2z = −3
4x − y + 8z = k
A. 11
B. 12
C. 13
D. 14
E. 15
D. 450
E. 400
D. −3
E. −5
Question 16
Consider the following macroeconomic model:
C = 0.7Y + 100
G = 50
I = 30
X = 80
M = 0.1Y + 50
Determine the equilibrium level of national income.
A. 525
B. 675
C. 550
Question 17
Solve the following system of equations to find z.
3x + 5y − z = −12
2x − 3y + 2z = 22
x + 4y − 3z = −10
A. 4
B. 2
C. 1
Question 18
Describe what happens to a good when the price of a complementary good decreases.
A. The equilibrium price increases, the equilibrium quantity decreases and the demand
curve shifts left.
B. The equilibrium price and quantity both decrease and the demand curve shifts left.
C. The equilibrium price and quantity both increase and the demand curve shifts right.
D. The equilibrium price decreases, the equilibrium quantity increases and the demand
curve shifts left.
E. The equilibrium price increases, the equilibrium quantity decreases and the demand
curve shifts right.
Examination questions
Question 1
Two take-away delivery services operate in the same town. EatMeNow charges a flat fee
of $9 for collection and an additional $0.50 per mile. Deliver4U does not have any flat fee
but charges $1.25 per mile.
(a) Draw graphs of the delivery charges against distance for both companies on the same
diagram, with 0 to 20 miles plotted on the horizontal axis and 0 to $25 plotted on the
vertical axis.
(b) Use your diagram to find the distance when both companies charge the same amount.
(c) If x and y denote distance and charge respectively, write down mathematical expres-
sions for each of the charge functions plotted in part (a) and use algebra to confirm that
your answer to part (b) is correct.
Question 2
The market demand and supply functions of a good are given by:
QD = 400 − 15P
QS = −50 + 10P
where QD, QS and P denote quantity demanded, quantity supplied and price, respectively.
(a) Work out the values of QD and QS when P = 5, 10, 15, 20 and 25, and hence draw
accurate graphs of the demand and supply curves on the same diagram with Q plotted
on the horizontal axis and P plotted on the vertical axis.
(b) Hence, or otherwise, find the equilibrium price and quantity.
(c) For what values of P is there excess demand?
(d) The government decides to impose a fixed tax of $5 on each good. Describe the effect,
if any, on the demand and supply curves on your diagram and calculate the new equilibrium price and quantity.
Question 3
The market equilibrium price and quantity of a good may be found by solving the simultaneous equations:
Q + 3P = 48 (1)
Q − 2P = 30 (2)
(a) Explain which of these equations represents the supply curve and which represents the
demand curve.
Examination questions 117
(b) If the demand curve were to be drawn with Q on the horizontal axis and P on the ver-
tical axis, state the values of the slope and vertical intercept.
(c) Determine the equilibrium price and quantity algebraically.
Question 4
In a simple macroeconomic model, the value of national income Y may be found by solving the system:
G
T
I
C
= 250
= 50
= 100
= 0.75Yd + 150
(government expenditure)
(taxation)
(planned investment)
(consumption)
where disposable income Yd = Y − T.
(a) Calculate the equilibrium level of national income.
(b) Calculate the total increase in government expenditure and investment needed to
increase the equilibrium level of national income by 20.
Question 5
The output of a production process, Q, is given by the function
Q5
KL
K 1 2L
where K and L denote capital and labour.
(a) Calculate the output when the capital and labour are 40 and 30 units, respectively.
(b) Find the level of labour needed to produce 4 units of output when 10 units of capital
are used.
(c) Transpose this formula to express K in terms of Q and L.
Question 6
The equations defining a simple model of the economy are:
C = 0.75Y + 18
I = 22
where C, Y and I denote consumption, national income and planned investment,
respectively.
Use this model to complete the following table of values and hence draw an accurate
graph of C + I plotted against Y on the interval 0 ≤ Y ≤ 200.
Y
0
100
200
C+I
By drawing a second line on the diagram, calculate the equilibrium level of national income.
Use your graph to explain what happens to the equilibrium level of national income
when the marginal propensity to consume decreases with autonomous consumption fixed.
118 Chapter 1 Linear Equations
Question 7
The demand and supply functions for two interdependent commodities are given by:
QD = 120 − 2P1 + P2; QS = −7 + 7P1
1
1
QD = 168 + 3P1 − 7P2; QS = −3 + 20P2
2
2
(a) Explain whether these goods are complementary or substitutable.
(b) Determine the equilibrium price and quantity.
(c) The government decides to provide a subsidy of $4 on each unit of good 1 only.
What effect does this subsidy have on the equilibrium prices of each good?
Question 8
(a) A baker has $3000 a month to spend on flour and sugar. Flour costs $3 per kilo and
sugar costs $4 per kilo. If the baker buys f kg of flour and s kg sugar, write down an
inequality for the budget constraint. Hence find the maximum amount of flour that
could be bought if the baker orders 270 kg of sugar.
(b) The demand function of a good is given by P 5
36
2Q 1 5
(i) Find P when Q = 3.5
(ii) Find Q when P = 2.4
(iii) Rearrange this formula to express Q in terms of P.
(c) Write down a simplified expression for the savings function given that the consumpY 2 1 100
tion function is C 5
.
Y12
Question 9
(a) Use the elimination method to solve the following system of equations expressing x
and y in terms of parameters m and n.
2x + 6y = 1
3x + my = n
For what value of m does this system fail to possess a unique solution?
What can you deduce about the solution when m takes this particular value?
Distinguish between the case when (i) n = 1.5 (ii) n ≠ 1.5.
(b) The equilibrium prices of three related goods satisfy the simultaneous equations
5P1 − 3P2 = 39
2P1 + 4P2 − 3P3 = 28
3P2 + P3 = 29
Use the elimination method to find P1, P2 and P3.
Examination questions 119
Question 10
Determine the equilibrium income Y and interest rate r, given the following information
about the commodity market:
C = 0.6Y + 60
I = −40r + 1300
where C and I denote consumption and planned investment, respectively, and the following information about the money market:
MS = 600
L1 = 0.2Y
L2 = −30r + 40
where MS, L1 and L2 denote money supply, transaction-precautionary demand for money
and speculative demand for money, respectively.
The marginal propensity to consume decreases. By producing a rough sketch of the IS
and LM curves on the same diagram, explain whether the equilibrium values of Y and r
will increase, decrease or stay the same.
Chapter 2
Non-linear Equations
The main aim of this chapter is to describe the mathematics of non-linear equations. The approach
is similar to that of Chapter 1. There are four sections. Section 2.1 should be read before
Section 2.2, and Section 2.3 should be read before Section 2.4.
The first section investigates the simplest non-linear equation, known as a quadratic. A quadratic
equation can easily be solved either by factorising it as the product of two linear factors or by
using a special formula. You are also shown how to sketch the graphs of quadratic functions.
The techniques are illustrated by finding the equilibrium price and quantity for quadratic supply
and demand functions.
Section 2.2 introduces additional functions in microeconomics, including revenue and profit.
There is very little new material in this section. It mainly consists of applying the ideas of
Section 2.1 to sketch graphs of quadratic revenue and profit functions and to find their
maximum values.
Finally, the topic of algebra, which we started in Chapter 1, is completed by investigating the
rules of indices and logarithms. The basic concepts are covered in Section 2.3. The notation
and rules of indices are extremely important and are used frequently in subsequent chapters.
Section 2.4 focuses on two specific functions, namely the exponential and natural logarithm
functions. If you run into difficulty, or are short of time, then this section could be omitted for the
time being, particularly if you do not intend to study the next chapter on the mathematics of
finance.
Section 2.1
Quadratic functions
Objectives
At the end of this section you should be able to:
●
Solve a quadratic equation using ‘the formula’.
●
Solve a quadratic equation given its factorisation.
●
Sketch the graph of a quadratic function using a table of function values.
●
Sketch the graph of a quadratic function by finding the coordinates of the intercepts.
●
Solve quadratic inequalities using graphs.
●
Solve inequalities using sign diagrams.
●
Determine equilibrium price and quantity given a pair of quadratic demand and supply
functions.
The first chapter considered the topic of linear mathematics. In particular, we described
how to sketch the graph of a linear function and how to solve a linear equation (or system of
simultaneous linear equations). It was also pointed out that not all economic functions are
of this simple form. In assuming that the demand and supply graphs are straight lines, we
are certainly making the mathematical analysis easy, but we may well be sacrificing realism.
It may be that the demand and supply graphs are curved and, in these circumstances, it is
essential to model them using more complicated functions. The simplest non-linear function
is known as a quadratic function and takes the form
f(x) = ax2 + bx + c
for some parameters a, b and c. (In fact, even if the demand function is linear, functions
derived from it, such as total revenue and profit, turn out to be quadratic. We investigate these
functions in the next section.) For the moment we concentrate on the mathematics of quadratics
and show how to sketch graphs of quadratic functions and how to solve quadratic equations.
Consider the elementary equation
x2 is an abbreviation
for x 3 x
x2 − 9 = 0
It is easy to see that the expression on the left-hand side is a special case of the above with
a = 1, b = 0 and c = −9. To solve this equation, we add 9 to both sides to get
x2 = 9
so we need to find a number, x, which when multiplied by itself produces the value 9. A
moment’s thought should convince you that there are exactly two numbers that work,
namely 3 and −3, because
3×3=9
and
(−3) × (−3) = 9
Section 2.1 Quadratic functions 123
These two solutions are called the square roots of 9. The symbol
is reserved for
the positive square root, so in this notation the solutions are 9 and 2 9 . These are usually
combined and written as 1
2 9 . The equation
x2 − 9 = 0
is trivial to solve because the number 9 has obvious square roots. In general, it is necessary
to use a calculator to evaluate square roots. For example, the equation
x2 − 2 = 0
can be written as
x2 = 2
and so has solutions x 5 6 2 . My calculator gives 1.414 213 56 (correct to eight decimal
places) for the square root of 2, so the above equation has solutions
1.414 213 56
and
−1.414 213 56
Example
Solve the following quadratic equations:
(a) 5x2 − 80 = 0 (b) x2 + 64 = 0 (c) (x + 4)2 = 81
Solution
(a) 5x2 − 80 = 0
5x2 = 80
(add 80 to both sides)
x = 16
(divide both sides by 5)
x = ±4
(square root both sides)
2
(b) x2 + 64 = 0
x2 = −64
(subtract 64 from both sides)
This equation does not have a solution because you cannot square a real number and
get a negative answer.
(c) (x + 4)2 = 81
x + 4 = ±9 (square root both sides)
The two solutions are obtained by taking the + and − signs separately. Taking the + sign,
x + 4 = 9 so
x=9−4=5
Taking the – sign,
x + 4 = −9 so x = −9 − 4 = −13
The two solutions are 5 and −13.
124 Chapter 2 Non-linear Equations
Practice Problem
1. Solve the following quadratic equations. (Round your solutions to two decimal places
if necessary.)
(a) x2 − 100 = 0
(b) 2x2 − 8 = 0
(c) x2 − 3 = 0
(e) x2 + 1 = 0
(f) 3x2 + 6.21 = 0
(g) x2 = 0
(d) x2 − 5.72 = 0
All of the equations considered in Practice Problem 1 are of the special form
ax2 + c = 0
in which the coefficient of x is zero. To solve more general quadratic equations, we use a
formula that enables the solutions to be calculated in a few lines of working. It can be shown
that
ax2 + bx + c = 0
has solutions
x5
2b 6 (b 2 2 4ac)
2a
The following example describes how to use this formula. It also illustrates the fact (which
you have already discovered in Practice Problem 1) that a quadratic equation can have two
solutions, one solution or no solutions.
Example
Solve the quadratic equations
(a) 2x2 + 9x + 5 = 0
(b) x2 − 4x + 4 = 0
(c) 3x2 − 5x + 6 = 0
Solution
(a) For the equation
2x2 + 9x + 5 = 0
we have a = 2, b = 9 and c = 5. Substituting these values into the formula
x5
2b 6 (b 2 2 4ac)
2a
gives
x5
29 6 (92 2 4(2)(5))
2(2)
5
29 6 (81 2 40)
4
5
29 6 41
4
Section 2.1 Quadratic functions 125
The two solutions are obtained by taking the + and − signs separately: that is,
29 1 41
5 20.649 (correct to three decimal places)
4
29 2 41
5 23.851 (correct to three decimal places)
4
It is easy to check that these are solutions by substituting them into the original
equation. For example, putting x = −0.649 into
2x2 + 9x + 5
gives
2(−0.649)2 + 9(−0.649) + 5 = 0.001 402
which is close to zero, as required. We cannot expect to produce an exact value of zero
because we rounded 41 to three decimal places. You might like to check for yourself
that −3.851 is also a solution.
(b) For the equation
x2 − 4x + 4 = 0
we have a = 1, b = −4 and c = 4. Substituting these values into the formula
x5
2b 6 (b 2 2 4ac)
2a
gives
x5
5
2(24) 6 ((24)2 2 4 (1)(4))
2(1)
4 6 (16 2 16)
2
46 0
2
460
5
2
5
Clearly we get the same answer irrespective of whether we take the + or the − sign
here. In other words, this equation has only one solution, x = 2. As a check, substitution of
x = 2 into the original equation gives
(2)2 − 4(2) + 4 = 0
(c) For the equation
3x2 − 5x + 6 = 0
we have a = 3, b = −5 and c = 6. Substituting these values into the formula
x5
2b 6 (b 2 2 4ac)
2a
➜
126 Chapter 2 Non-linear Equations
gives
x5
2(25) 6 ((25)2 2 4(3)(6))
2(3)
5
5 6 (25 2 72)
6
5
5 6 (247)
6
The number under the square root sign is negative and, as you discovered in Practice
Problem 1, it is impossible to find the square root of a negative number. We conclude
that the quadratic equation
3x2 − 5x + 6 = 0
has no solutions.
This example demonstrates the three cases that can occur when solving quadratic equations.
The precise number of solutions that an equation can have depends on whether the number under
the square root sign is positive, zero or negative. The number b2 − 4ac is called the discriminant
because the sign of this number discriminates between the three cases that can occur.
●
If b2 − 4ac > 0, then there are two solutions:
x5
●
and x 5
2b 2 (b 2 2 4ac)
2a
If b2 − 4ac = 0, then there is one solution:
x5
●
2b 1 (b 2 2 4ac)
2a
2b 6 0
2b
5
2a
2a
If b2 − 4ac < 0, then there are no solutions because (b 2 2 4ac) does not exist.
Practice Problem
2. Solve the following quadratic equations (where possible):
(a) 2x2 − 19x − 10 = 0
(b) 4x2 + 12x + 9 = 0
(c) x2 + x + 1 = 0
(d) x2 − 3x + 10 = 2x + 4
The following example demonstrates an interesting application in business.
SectiOn 2.1 QuADrATIC FuNCTIoNS
127
example
A company sells packs of printer ink cartridges for $24. If a customer orders more than
100 packs, the company is prepared to reduce the unit price by 4 cents for each pack
bought above 100 up to a maximum of 300 in a single order.
(a) How much does it cost to buy 130 packs?
(b) If the cost is $5324, how many packs are ordered?
Solution
(a) The first 100 packs cost $24 each, so total cost for these is 100 × 24 = $2400.
The unit cost of the remaining 30 packs is 24 − 0.04 × 30 = $22.80, so the total cost
of the extra 30 packs is 30 × 22.80 = $684. The cost of the complete order is $3084.
(b) The total cost of the packs bought in excess of 100 is 5324 − 2400 = $2924.
If x denotes the number of packs above 100, then the unit price of each of these is
24 − 0.04x, giving a total cost of (24 − 0.04x)x = 24x − 0.04x2.
Hence
24x − 0.04x2 = 2924
which rearranges as 0.04x2 − 24x + 2924 = 0.
This quadratic equation can then be solved using the formula:
x5
5
2(224) 6 (224)2 2 4(0.04)(2924)
24 6 576 2 467.84
5
0.08
2(0.04)
24 6 108.16
24 6 10.4
5
0.08
0.08
giving two solutions, x = 170 and x = 430. The maximum permissible order is 300, so
x = 170 giving a total order of 100 + 170 = 270 packs of cartridges.
You may be familiar with another method for solving quadratic equations. This is based
on the factorisation of a quadratic into the product of two linear factors. Section 1.1
described how to multiply out two brackets. One of the examples in that section showed that
(x + 1)(x + 2) = x2 + 3x + 2
Consequently, the solutions of the equation
x2 + 3x + 2 = 0
are the same as those of
(x + 1)(x + 2) = 0
Now the only way that two numbers can be multiplied together to produce a value of zero is
when (at least) one of the numbers is zero.
if ab = 0, then either a = 0 or b = 0 (or both)
It follows that either
x + 1 = 0 with solution
x = −1
128 Chapter 2 Non-linear Equations
or
x+2=0
with solution x = −2
The quadratic equation
x2 + 3x + 2 = 0
therefore has two solutions, x = −1 and x = −2.
The difficulty with this approach is that it is impossible, except in very simple cases, to
work out the factorisation from any given quadratic, so the preferred method is to use the
formula. However, if you are lucky enough to be given the factorisation, or perhaps clever
enough to spot the factorisation for yourself, then it does provide a viable alternative.
One important feature of linear functions is that their graphs are always straight lines.
Obviously, the intercept and slope vary from function to function, but the shape is always
the same. It turns out that a similar property holds for quadratic functions. Now, whenever
you are asked to produce a graph of an unfamiliar function, it is often a good idea to tabulate
the function, to plot these points on graph paper and to join them up with a smooth curve.
The precise number of points to be taken depends on the function but, as a general rule,
between 5 and 10 points usually produce a good picture.
A table of values for the simple square function
f (x) = x2
is given by
x
f(x)
−3
−2
−1
0
1
2
3
9
4
1
0
1
4
9
The first row of the table gives a selection of ‘incoming’ numbers, x, while the second row
shows the corresponding ‘outgoing’ numbers, y. Points with coordinates (x, y) are then
plotted on graph paper to produce the curve shown in Figure 2.1. For convenience, different
scales are used on the x and y axes.
Mathematicians call this curve a parabola, whereas economists refer to it as U-shaped.
Notice that the graph is symmetric about the y axis with a minimum point at the origin; if a
mirror is placed along the y axis, then the left-hand part is the image of the right-hand part.
y
10
y 5 x2
8
6
4
2
23
Figure 2.1
22
21
1
2
3
x
Section 2.1 Quadratic functions 129
Advice
The following problem is designed to give you an opportunity to tabulate and sketch graphs of more
general quadratic functions. Please remember that when you substitute numbers into a formula, you
must use BIDMAS to decide the order of the operations. For example, in part (a) you need to substitute x = −1 into 4x2 − 12x + 5. You get
4(−1)2 − 12(−1) + 5
= 4 + 12 + 5
= 21
Note also that when using a calculator, you must use brackets when squaring negative numbers. In
this case a possible sequence of key presses might be
4
(
(2)
1
)
x2
12
2
3
(2)
1
1
5
5
Practice Problem
3. Complete the following tables of function values and hence sketch a graph of each
quadratic function.
(a) f (x) = 4x2 − 12x + 5
x
−1
0
1
2
3
4
f(x)
(b) f (x) = −x2 + 6x − 9
x
0
1
2
3
4
5
6
f(x)
(c) f (x) = −2x2 + 4x − 6
x
−2
−1
0
1
2
3
4
f(x)
The results of Practice Problem 3 suggest that the graph of a quadratic is always parabolic.
Furthermore, whenever the coefficient of x2 is positive, the graph bends upwards and is a
‘happy’ parabola (U shape). A selection of U-shaped curves is shown in Figure 2.2. Similarly,
when the coefficient of x2 is negative, the graph bends downwards and is a ‘sad’ parabola
(inverted U shape). A selection of inverted U-shaped curves is shown in Figure 2.3.
The task of sketching graphs from a table of function values is extremely tedious, particularly if only a rough sketch is required. It is usually more convenient just to determine a few
key points on the curve. The obvious points to find are the intercepts with the coordinate
axes, since these enable us to ‘tether’ the parabola down in the various positions shown in
Figures 2.2 and 2.3. The curve crosses the y axis when x = 0. Evaluating the function
f (x) = ax2 + bx + c
at x = 0 gives
f (0) = a(0)2 + b(0) + c = c
130 Chapter 2 Non-linear Equations
y
y
A
A
B
B
x
x
(a)
(a)
y
y
x
x
(b)
(b)
y
y
x
(c)
x
(c)
Figure 2.2
Figure 2.3
so the constant term determines where the curve cuts the vertical axis (as it did for linear
functions). The curve crosses the x axis when y = 0 or, equivalently, when f(x) = 0, so we
need to solve the quadratic equation
ax2 + bx + c = 0
This can be done using ‘the formula’ and the solutions are the points where the graph cuts
the horizontal axis. In general, a quadratic equation can have two, one or no solutions and
these possibilities are illustrated in cases (a), (b) and (c), respectively, in Figures 2.2 and 2.3.
In case (a) the curve crosses the x axis at A, turns round and crosses it again at B, so there are
two solutions. In case (b) the curve turns round just as it touches the x axis, so there is only
one solution. Finally, in case (c) the curve turns round before it has a chance to cross the x
axis, so there are no solutions.
The strategy for sketching the graph of a quadratic function
f (x) = ax2 + bx + c
may now be stated.
Step 1
Determine the basic shape. The graph has a U shape if a > 0, and an inverted U shape if a < 0.
Section 2.1 Quadratic functions 131
Step 2
Determine the y intercept. This is obtained by substituting x = 0 into the function, which
gives y = c.
Step 3
Determine the x intercepts (if any). These are obtained by solving the quadratic equation
ax2 + bx + c = 0
This three-step strategy is illustrated in the following example.
Example
Give a rough sketch of the graph of the following quadratic function:
f (x) = −x2 + 8x − 12
Solution
For the function
f (x) = −x2 + 8x − 12
the strategy is as follows.
Step 1
The coefficient of x2 is −1, which is negative, so the graph is a ‘sad’ parabola with an
inverted U shape.
Step 2
The constant term is −12, so the graph crosses the vertical axis at y = −12.
Step 3
For the quadratic equation
−x2 + 8x − 12 = 0
the formula gives
x5
5
28 6 (64 2 48)
28 6 (82 2 4(21)(212))
5
2(21)
22
28 6 16
28 6 4
5
22
22
so the graph crosses the horizontal axis at
x5
28 1 4
52
22
and
x5
28 2 4
56
22
The information obtained in steps 1–3 is sufficient to produce the sketch shown in
Figure 2.4.
➜
132 Chapter 2 Non-linear Equations
y
maximum point
4
4
2
6
x
y 5 2x2 1 8x 2 12
212
Figure 2.4
In fact, we can go even further in this case and locate the coordinates of the turning
point – that is, the maximum point – on the curve. By symmetry, the x coordinate of this
point occurs exactly halfway between x = 2 and x = 6: that is, at
x = 1/2 (2 + 6) = 4
The corresponding y coordinate is found by substituting x = 4 into the function to get
f (4) = −(4)2 + 8(4) − 12 = 4
The maximum point on the curve therefore has coordinates (4, 4).
Practice Problem
4. Use the three-step strategy to produce rough graphs of the following quadratic functions:
(a) f (x) = 2x2 − 11x − 6 (b) f(x) = x2 − 6x + 9
One useful by-product of our work on sketching graphs is that it enables us to solve quad­
ratic inequalities with no extra effort.
Example
Solve the following quadratic inequalities:
(a) −x2 + 8x − 12 > 0 (b) −x2 + 8x − 12 ≤ 0
Section 2.1 Quadratic functions 133
Solution
The graph of the function f (x) = −x2 + 8x − 12 has already been sketched in Figure 2.4.
The parabola lies above the x axis (that is, the line y = 0) between 2 and 6 and is below
the x axis outside these values.
(a) The quadratic function takes positive values when the graph is above the x axis, so
the inequality has the solution 2 < x < 6. The values of 2 and 6 must be excluded from
the solution since we require the quadratic to be strictly greater than zero.
(b) The graph is on or below the x axis at or to the left of 2, and at or to the right of 6, so
the complete solution is x ≤ 2 and x ≥ 6.
Practice Problem
5. Use your answers to Practice Problem 4 to write down the solution to each of the
following quadratic inequalities:
(a) 2x2 − 11x − 6 ≤ 0 (b) x2 − 6x + 9 > 0
If the quadratic is in factorised form, an alternative method can be used to solve the associated inequality. This is based on a sign diagram. It avoids the need to draw a graph and the
method has the added advantage that it can be used to solve other inequalities. We illustrate
the technique in the following example.
Example
Use a sign diagram to solve the following inequalities:
(a) (x − 2)(x + 3) ≥ 0 (b)
x
,0
x12
Solution
(a) We know that the factor x − 2 is zero at x = 2. If x is smaller than 2, the factor is
negative (for example, when x = 1, the factor takes the value −1 < 0), and when x is
bigger than 2, the factor is positive (for example, when x = 4, the factor takes the value
2 > 0). These results are illustrated on the number line:
(x 2 2) 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 1 1 1 1
25
24
23
22
21
0
1
2
3
4
The second factor, x + 3, takes the value zero at x = −3, is negative to the left of −3
and is positive to the right of −3. This is illustrated in the number line diagram:
(x 1 3) 2 2 2 2 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1
25
24
23
22
21
0
1
2
3
4
➜
134 Chapter 2 Non-linear Equations
The expression (x − 2)(x + 3) is the product of the two factors. To the left of −3 the
number lines show that both factors are negative, so their product is positive. Between
−3 and 2 one factor is negative and the other positive, so their product is negative. Of
course, if one factor is zero, the product is automatically zero, irrespective of the sign
of the second factor. The complete sign diagram for the product is shown below.
(x 2 2)(x 1 3) 1 1 1 1 0 2 2 2 2 2 2 2 2 2 0 1 1 1 1
25
24
23
22
21
0
1
2
3
4
The diagram shows immediately that the inequality (x − 2)(x + 3) ≥ 0 is satisfied by
x ≤ −3, x ≥ 2.
(b) The factor (x + 2) is zero at x = −2, negative to the left of x = −2, and positive to the
right of this. The factor x is obviously zero at x = 0, negative to the left of x = 0, and
positive to the right of this. The complete sign diagram is shown below:
x
2 2 2 2 2 2 2 2 2 2 0 1 1 1 1
25
24
23
22
21
0
1
2
(x 1 2) 2 2 2 2 2 2 0 1 1 1 1 1 1 1 1
25
24
23
22
21
0
1
2
x/(x 1 2) 1 1 1 1 1 1 ` 2 2 2 0 1 1 1 1
25
24
23
22
21
0
1
2
The rules for dividing negative numbers are the same as those for multiplying, so the
diagram is completed in the same way as before. The only exception occurs at x = −2
because we cannot divide by zero. This is indicated on the diagram by putting the
symbol ∞ (infinity) at this position on the line. This diagram shows that the inequality
x
, 0 is satisfied by −2 < x < 0.
x12
Practice Problem
6. Use a sign diagram to solve the following inequalities:
(a) (x − 1)(x − 4) ≤ 0 (b)
x 21
$0
x12
We conclude this section by solving a particular problem in microeconomics. In Section 1.5
the concept of market equilibrium was introduced, and in each of the problems the supply
and demand functions were always given to be linear. The following example shows this to
be an unnecessary restriction and indicates that it is almost as easy to manipulate quadratic
supply and demand functions.
Section 2.1 Quadratic functions 135
Example
Given the supply and demand functions
P = Q2S + 14QS + 22
P = −QD2 − 10QD + 150
calculate the equilibrium price and quantity.
Solution
In equilibrium, QS = QD, so if we denote this equilibrium quantity by Q, the supply and
demand functions become
P = Q2 + 14Q + 22
P = −Q2 − 10Q + 150
Hence
Q2 + 14Q + 22 = −Q2 − 10Q + 150
since both sides are equal to P. Collecting like terms gives
2Q2 + 24Q − 128 = 0
which is just a quadratic equation in the variable Q. Before using the formula to solve this
it is a good idea to divide both sides by 2 to avoid large numbers. This gives
Q2 + 12Q − 64 = 0
and so
Q5
212 6 ((122) 2 4(1)(264))
2(1)
212 6 (400)
2
212 6 20
5
2
5
The quadratic equation has solutions Q = −16 and Q = 4. Now, the solution Q = −16
can obviously be ignored because a negative quantity does not make sense. The equilibrium quantity is therefore 4. The equilibrium price can be calculated by substituting this
value into either the original supply or demand equation.
From the supply equation,
P = 42 + 14(4) + 22 = 94
As a check, the demand equation gives
P = −(4)2 − 10(4) + 150 = 94
✓
➜
136 Chapter 2 Non-linear Equations
You might be puzzled by the fact that we actually obtain two possible solutions, one
of which does not make economic sense. The supply and demand curves are sketched in
Figure 2.5. This shows that there are indeed two points of intersection, confirming the
mathematical solution. However, in economics the quantity and price are both positive, so
the functions are defined only in the top right-hand (that is, positive) quadrant. In this
region there is just one point of intersection, at (4, 94).
P
supply
(4, 94)
(216, 54)
Q
demand
Figure 2.5
Practice Problem
7. Given the supply and demand functions
P = 2QS2 + 10QS + 10
P = −QD2 − 5QD + 52
calculate the equilibrium price and quantity.
Key Terms
Discriminant The number b2 − 4ac, which is used to indicate the number of solutions of the
quadratic equation ax2 + bx + c = 0.
Parabola
The shape of the graph of a quadratic function.
Quadratic function A function of the form f(x) = ax2 + bx + c where a ≠ 0.
Square root A number that when multiplied by itself equals a given number; the solutions of the
equation x2 = c which are written 6 c .
U-shaped curve A term used by economists to describe a curve, such as a parabola, which bends
upwards, like the letter U.
Section 2.1 Quadratic functions 137
Exercise 2.1
1. Solve the following quadratic equations:
(a) x2 = 81
(b) x2 = 36
(c) 2x2 = 8
(d) (x − 1)2 = 9
(e) (x + 5)2 = 16
2. Write down the solutions of the following equations:
(a) (x − 1)(x + 3) = 0
(b) (2x − 1)(x + 10) = 0
(d) (3x + 5)(4x − 9) = 0
(e) (5 − 4x)(x − 5) = 0
(c) x(x + 5) = 0
3. Use ‘the formula’ to solve the following quadratic equations. (Round your answers to
two decimal places.)
(a) x2 − 5x + 2 = 0
(b) 2x2 + 5x + 1 = 0
(c) −3x2 + 7x + 2 = 0
(d) x2 − 3x − 1 = 0
(e) 2x2 + 8x + 8 = 0
(f) x2 − 6x + 10 = 0
4. Solve the equation f (x) = 0 for each of the following quadratic functions:
(a) f (x) = x2 − 16
(b) f(x) = x(100 − x)
(c) f(x) = −x2 + 22x − 85
(d) f (x) = x2 − 18x + 81
(e) f(x) = 2x2 + 4x + 3
5. Sketch the graphs of the quadratic functions given in Question 4.
6. Use the results of Question 5 to solve each of the following inequalities:
(a) x2 − 16 ≥ 0
(b) x(100 − x) > 0
(d) x − 18x + 81 ≤ 0
(e) 2x + 4x + 3 > 0
2
(c) −x2 + 22x − 85 ≥ 0
2
7. The production levels of coffee in Mexico, Q (in suitable units) depends on the average
summer temperature, T (in °C).
A statistical model of recent data shows that Q = −0.046T 2 + 2.3T + 27.6.
(a) Complete the table of values and draw a graph of Q against T in the range, 23 ≤ T ≤ 30.
T
23
24
25
26
27
28
29
30
Q
(b) Average summer temperatures over the last few decades have been about 25°C.
However, some climate change models predict that this could rise by several degrees
over the next 50 years. Use your graph to comment on the likely impact that this
may have on coffee growers in Mexico.
8. Use a sign diagram to solve the following inequalities:
(a) x(x − 3) > 0
(b) (x − 1)(x + 1) ≥ 0
(c)
x14
,0
x22
9. Given the quadratic supply and demand functions
P = Q2S + 2QS + 12
P = −QD2 − 4QD + 68
determine the equilibrium price and quantity.
➜
138 Chapter 2 Non-linear Equations
10. Given the supply and demand functions
P = Q2S + 2QS + 7
P = −QD + 25
determine the equilibrium price and quantity.
11. A clothing supplier sells T-shirts to retailers for $7 each. If a store agrees to buy more
than 30, the supplier is willing to reduce the unit price by 3 cents for each shirt bought
above 30, with a maximum single order of 100 shirts.
(a) How much does an order of 40 shirts cost?
(b) If the total cost of an order is $504.25, how many T-shirts did the store buy
altogether?
Exercise 2.1*
1. Solve the following quadratic equations:
(a) x2 = 169
(b) (x − 5)2 = 64
(c) (2x − 7)2 = 121
2. Find the solutions (in terms of d) of the quadratic equation
x2 + 6dx − 7d 2 = 0
3. Write down the solutions of the following equations:
(a) (x − 3)(x + 8) = 0
(b) (3x − 2)(2x + 9) = 0
(d) (6x − 1) = 0
(e) (x − 2)(x + 1)(4 − x) = 0
2
(c) x(4x − 3) = 0
4. Solve the following quadratic equations, rounding your answers to two decimal places,
if necessary:
(a) x2 − 15x + 56 = 0
(b) 2x2 − 5x + 1 = 0
(c) 4x2 − 36 = 0
(d) x2 − 14x + 49 = 0
(e) 3x2 + 4x + 7 = 0
(f) x2 − 13x + 200 = 16x + 10
5. Solve the following inequalities:
(a) x2 ≥ 64
(b) x2 − 10x + 9 ≤ 0
(d) −3x2 + 2x + 5 ≥ 0
(e) x2 + 2x + 1 ≤ 0
6. One solution of the quadratic equation
x2 − 8x + c = 0
is known to be x = 2. Find the second solution.
7. Find the value of k so that the equation
x2 − 10x + 2k = 8x − k
has exactly one root.
(c) 2x2 + 15x + 7 < 0
Section 2.1 Quadratic functions 139
8. Use a sign diagram to solve the following inequalities:
(a) (x + 3)(x − 4) ≥ 0
(x 2 2)
(d)
$0
(x 2 3)(x 2 5)
(b) (2 − x)(x + 1) > 0
(c) (x − 1)(x − 2)(x − 3) ≤ 0
9. A firm’s monthly cost for paying cleaners’ wages is $47 250. Under a new pay deal each
cleaner earns $375 more each month. If the new pay deal goes through, the firm realises
that it will need to reduce the number of cleaners by 3 if it is to cover its costs within
the existing budget. What is the monthly salary of a cleaner before the pay rise?
10. Given the supply and demand functions
P = Q2S + 10QS + 30
P = −QD2 − 8QD + 200
calculate the equilibrium price, correct to two decimal places.
11. A pottery can make B bowls and P plates in a week according to the relation
2B2 + 5B + 25P = 525
(a) If it makes five bowls, how many plates can it make in a week?
(b) What is the maximum number of bowls that it can produce in a week?
12. A city centre tour guide currently charges $34 for a full day’s tour. The average number
of customers is 48. Market research suggests that for every $1 increase in tour price, the
guide can expect to lose two customers per tour.
(a) Show that if the price increase is $x, then the expected revenue from each tour is
−2x2 − 20x + 1632
(b) The guide needs to ensure that the expected revenue is at least $1440. By solving a
quadratic inequality, find the range of prices that need to be charged.
(c) What price should be charged to maximise expected revenue?
13. Given the supply and demand functions
QS 5 (P 1 8) P 1 20
QD 5
460 2 12P 2 3P2
P 1 20
calculate the equilibrium price and quantity.
Section 2.2
Revenue, cost and profit
Objectives
At the end of this section you should be able to:
●
Sketch the graphs of the total revenue, total cost, average cost and profit functions.
●
Find the level of output that maximises total revenue.
●
Find the level of output that maximises profit.
●
Find the break-even levels of output.
The main aim of this section is to investigate one particular function in economics, namely
profit. By making reasonable simplifying assumptions, the profit function is shown to be
quadratic and so the methods developed in Section 2.1 can be used to analyse its properties.
We describe how to find the levels of output required for a firm to break even and to maximise profit. The profit function is denoted by the Greek letter π (pi, pronounced ‘pie’) and is
defined to be the difference between total revenue, TR, and total cost, TC: that is,
π = TR − TC
This definition is entirely sensible because TR is the amount of money received by the firm
from the sale of its goods, and TC is the amount of money that the firm has to spend to produce these goods. We begin by considering the total revenue and total cost functions in turn.
The total revenue received from the sale of Q goods at price P is given by
TR = PQ
For example, if the price of each good is $70 and the firm sells 300, then the revenue is
$70 × 300 = $21 000
Given any particular demand function, expressing P in terms of Q, it is a simple matter to
obtain a formula for TR solely in terms of Q. A graph of TR against Q can then be sketched.
Example
Given the demand function
P = 100 − 2Q
express TR as a function of Q and hence sketch its graph.
(a) For what values of Q is TR zero?
(b) What is the maximum value of TR?
Solution
Total revenue is defined by
TR = PQ
Section 2.2 Revenue, cost and profit 141
and, since P = 100 − 2Q, we have
TR = (100 − 2Q)Q = 100Q − 2Q2
This function is quadratic and so its graph can be sketched using the strategy described in
Section 2.1.
Step 1
The coefficient of Q2 is negative, so the graph has an inverted U shape.
Step 2
The constant term is zero, so the graph crosses the TR axis at the origin.
Step 3
To find where the curve crosses the horizontal axis, we could use ‘the formula’. However,
this is not necessary, since it follows immediately from the factorisation
TR = (100 − 2Q)Q
that TR = 0 when either 100 − 2Q = 0 or Q = 0. In other words, the quadratic equation
has two solutions, Q = 0 and Q = 50.
The total revenue curve is shown in Figure 2.6.
(a) From Figure 2.6, the total revenue is zero when Q = 0 and Q = 50.
(b) By symmetry, the parabola reaches its maximum halfway between 0 and 50: that is, at
Q = 25. The corresponding total revenue is given by
TR = 100(25) − 2(25)2 = 1250
TR
(25, 1250)
TR = 100Q 2 2Q2
0
50
Q
Figure 2.6
Practice Problem
1. Given the demand function
P = 1000 − Q
express TR as a function of Q and hence sketch a graph of TR against Q. What value
of Q maximises total revenue, and what is the corresponding price?
142 Chapter 2 Non-linear Equations
In general, given the linear demand function
P = aQ + b (a < 0, b > 0)
the total revenue function is
TR = PQ = (aQ + b)Q
= aQ2 + bQ
This function is quadratic in Q and, since a < 0, the TR curve has an inverted U shape.
Moreover, since the constant term is zero, the curve always intersects the vertical axis at the
origin. This fact should come as no surprise to you; if no goods are sold, the revenue must
be zero.
We now turn our attention to the total cost function, TC, which relates the production
costs to the level of output, Q. As the quantity produced rises, the corresponding cost also
rises, so the TC function is increasing. However, in the short run, some of these costs are
fixed. Fixed costs, FC, include the cost of land, equipment, rent and possibly skilled labour.
Obviously, in the long run all costs are variable, but these particular costs take time to vary,
so they can be thought of as fixed in the short run. Variable costs, on the other hand, vary with
output and include the cost of raw materials, components, energy and unskilled labour. If VC
denotes the variable cost per unit of output, then the total variable cost, TVC, in producing
Q goods is given by
TVC = (VC)Q
The total cost is the sum of the contributions from the fixed and variable costs, so is
given by
TC = FC + (VC)Q
Now, although this is an important economic function, it does not always convey the
information necessary to compare individual firms. For example, suppose that an inter­
national car company operates two plants, one in the United States and one in Europe, and
suppose that the total annual costs are known to be $200 million and $45 million, respectively. Which of these two plants is regarded as the more efficient? Unfortunately, unless we
also know the total number of cars produced, it is impossible to make any judgement. The
significant function here is not the total cost, but rather the average cost per car. If the US and
European plants manufacture 80 000 and 15 000 cars, respectively, their corresponding average costs are
200 000 000
5 2500
80 000
and
45 000 000
5 3000
15 000
On the basis of these figures, the US plant appears to be the more efficient. In practice, other
factors would need to be taken into account before deciding to increase or decrease the scale
of operation in either country.
In general, the average cost function, AC, is obtained by dividing the total cost by output,
so that
Section 2.2 Revenue, cost and profit 143
AC 5
TC
FC 1 (VC)Q
5
Q
Q
5
FC
(VC)Q
1
Q
Q
5
FC
1 VC
Q
If the fixed costs FC = 1000 and variable costs per unit VC = 4, then the expressions for the
total and average costs functions are
TC = 1000 + 4Q
and
AC 5
5
TC 1000 1 4Q
5
Q
Q
1000
14
Q
The graph of the total cost function is easily sketched. It is a straight line with intercept 1000
and slope 4. It is sketched in Figure 2.7. The average cost function is of a form that we have
not met before, so we have no prior knowledge about its basic shape. Under these circumstances
it is useful to tabulate the function. The tabulated values are then plotted on graph paper
and a smooth curve obtained by joining the points together. One particular table of function
values is
Q
100
250
500
1000
2000
AC
14
8
6
5
4.5
These values are readily checked. For instance, when Q = 100,
AC 5
1000
1 4 5 10 1 4 5 14
100
A graph of the average cost function, based on this table, is sketched in Figure 2.8. This
curve is known as a rectangular hyperbola and is sometimes referred to by economists as being
L-shaped.
TC
slope 5 VC 5 4
intercept 5 FC 5 1000
Q
Figure 2.7
144 Chapter 2 Non-linear Equations
AC
14
12
AC 5 1000
14
Q
10
8
6
4
2
500
1000
1500
2000
Q
Figure 2.8
Practice Problem
2. Given that fixed costs are 100 and that variable costs are 2 per unit, express TC and
AC as functions of Q. Hence sketch their graphs.
In general, whenever the variable cost, VC, is a constant, the total cost function,
TC = FC + (VC)Q
is linear. The intercept is FC and the slope is VC. For the average cost function
AC 5
FC
1 VC
Q
note that if Q is small, then FC/Q is large, so the graph bends sharply upwards as Q
approaches zero. As Q increases, FC/Q decreases and eventually tails off to zero for large
values of Q. The AC curve therefore flattens off and approaches VC as Q gets larger and
larger. This phenom­enon is hardly surprising, since the fixed costs are shared between more
and more goods, so have little effect on AC for large Q. The graph of AC therefore has the
basic L shape shown in Figure 2.9. This effect, in which the average cost decreases as the
number of units produced increases, is one reason for economies of scale, which encourage
the growth of large organisations. This discussion assumes that VC is a constant. In practice,
this may not be the case, and VC might depend on Q. The TC graph is then no longer linear
and the AC graph becomes U-shaped rather than L-shaped. Examples of this can be found in
Questions 5 and 11 in Exercise 2.2 at the end of this section.
Figure 2.10 shows typical TR and TC graphs sketched on the same diagram. These are
drawn on the assumption that the demand function is linear (which leads to a quadratic total
revenue function) and that the variable costs are constant (which leads to a linear total cost
function). The horizontal axis represents quantity, Q. Strictly speaking, the label Q means
different things for the two functions. For the revenue function, Q denotes the quantity
of goods actually sold, whereas for the cost function, it denotes the quantity produced. In
Section 2.2 Revenue, cost and profit 145
AC
graph ‘blows up’ as
Q approaches zero
graph approaches VC
as Q gets larger
VC
Q
Figure 2.9
total
revenue
maximum
profit
B
total
cost
A
QA
QB
Q
Figure 2.10
sketching both graphs on the same diagram, we are implicitly assuming that these two values
are the same and that the firm sells all of the goods that it produces.
The two curves intersect at precisely two points, A and B, corresponding to output levels
QA and QB. At these points the cost and revenue are equal and the firm breaks even. If Q < QA
or Q > QB, then the TC curve lies above that of TR, so cost exceeds revenue. For these levels of
output the firm makes a loss. If QA < Q < QB, then revenue exceeds cost and the firm makes a
profit which is equal to the vertical distance between the revenue and cost curves. The maximum profit occurs where the gap between them is largest. The easiest way of calculating maximum profit is to obtain a formula for profit directly in terms of Q using the defining equation
π = TR − TC
146 Chapter 2 Non-linear Equations
Example
If fixed costs are 4, variable costs per unit are 1 and the demand function is
P = 10 − 2Q
obtain an expression for π in terms of Q and hence sketch a graph of π against Q.
(a) For what values of Q does the firm break even?
(b) What is the maximum profit?
Solution
We begin by obtaining expressions for the total cost and total revenue. For this problem,
FC = 4 and VC = 1, so
TC = FC + (VC)Q = 4 + Q
The given demand function is P = 10 − 2Q
so TR = PQ = (10 − 2Q)Q = 10Q − 2Q2
Hence the profit is given by
π = TR − TC
= (10Q − 2Q2) − (4 + Q)
= 10Q − 2Q2 − 4 − Q
= −2Q2 + 9Q − 4
To sketch a graph of the profit function we follow the strategy described in Section 2.1.
Step 1
The coefficient of Q2 is negative, so the graph has an inverted U shape.
Step 2
The constant term is −4, so the graph crosses the vertical axis when π = −4.
Step 3
The graph crosses the horizontal axis when π = 0, so we need to solve the quadratic equation
−2Q2 + 9Q − 4 = 0
This can be done using ‘the formula’ to get
Q5
29 6 81 2 32 29 6 7
5
2(22)
24
so Q = 0.5 and Q = 4.
The profit curve is sketched in Figure 2.11.
(a) From Figure 2.11 we see that profit is zero when Q = 0.5 and Q = 4.
(b) By symmetry, the parabola reaches its maximum halfway between 0.5 and 4: that is, at
Q 5 1/2(0.5 1 4) 5 2.25
The corresponding profit is given by
π = −2(2.25)2 + 9(2.25) − 4 = 6.125
Section 2.2 Revenue, cost and profit 147
π
6.125
maximum
profit
0.5
2.25
4
Q
break-even points
24
Figure 2.11
Advice
It is important to notice the use of brackets in the previous derivation of π. A common student mistake is to forget to include the brackets and just to write down
π = TR − TC
= 10Q − 2Q2 − 4 + Q
= −2Q2 + 11Q − 4
This cannot be right, since the whole of the total cost needs to be subtracted from the total revenue,
not just the fixed costs. You might be surprised to learn that many economics students make this sort
of blunder, particularly under examination conditions.
Practice Problem
3. If fixed costs are 25, variable costs per unit are 2 and the demand function is
P = 20 − Q
obtain an expression for π in terms of Q and hence sketch its graph.
(a) Find the levels of output which give a profit of 31.
(b) Find the maximum profit and the value of Q at which it is achieved.
148 Chapter 2 Non-linear Equations
Key Terms
Total cost per unit of output: AC = TC/Q.
Average cost
Total costs that are independent of output.
Fixed costs
L-shaped curve
A term used by economists to describe the graph of a function, such as f(x) 5 a 1
which bends roughly like the letter L.
Profit
b
,
x
Total revenue minus total cost: π = TR − TC.
Rectangular hyperbola
A term used by mathematicians to describe the graph of a function, such
b
as f(x) 5 a 1 , which is a hyperbola with horizontal and vertical asymptotes.
x
Total cost
The sum of the total variable and fixed costs: TC = TVC + FC.
Total revenue
A firm’s total earnings from the sales of a good: TR = PQ.
Variable costs
Total costs that change according to the amount of output produced.
Exercise 2.2
1. (a) If the demand function of a good is given by
P = 80 − 3Q
find the price when Q = 10, and deduce the total revenue.
(b) If fixed costs are 100 and variable costs are 5 per unit, find the total cost when
Q = 10.
(c) Use your answers to parts (a) and (b) to work out the corresponding profit.
2. Given the following demand functions, express TR as a function of Q and hence sketch
the graphs of TR against Q:
(a) P = 4
(b) P = 7/Q
(c) P = 10 − 4Q
3. Given the following total revenue functions, find the corresponding demand functions:
(a) TR = 50Q − 4Q2
(b) TR = 10
4. Given that fixed costs are 500 and that variable costs are 10 per unit, express TC and
AC as functions of Q. Hence sketch their graphs.
5. Given that fixed costs are 1 and that variable costs are Q + 1 per unit, express TC and
AC as functions of Q. Hence sketch their graphs.
6. The total cost, TC, of producing 100 units of a good is 600 and the total cost of
producing 150 units is 850. Assuming that the total cost function is linear, find an
expression for TC in terms of Q, the number of units produced.
7. The total cost of producing 500 items a day in a factory is $40 000, which includes a
fixed cost of $2000.
(a) Work out the variable cost per item.
(b) Work out the total cost of producing 600 items a day.
Section 2.2 Revenue, cost and profit 149
8. A taxi firm charges a fixed cost of $10 together with a variable cost of $3 per mile.
(a) Work out the average cost per mile for a journey of 4 miles.
(b) Work out the minimum distance travelled if the average cost per mile is to be less
than $3.25.
9. Find an expression for the profit function given the demand function
2Q + P = 25
and the average cost function
AC 5
32
15
Q
Find the values of Q for which the firm
(a) breaks even;
(b) makes a loss of 432 units;
(c) maximises profit.
10. Sketch, on the same diagram, graphs of the total revenue and total cost functions,
TR = −2Q2 + 14Q
TC = 2Q + 10
(1) Use your graphs to estimate the values of Q for which the firm
(a) breaks even;
(b) maximises profit.
(2) Confirm your answers to part (1) using algebra.
11. The demand function for a firm’s product is given by P = 60 − Q.
Fixed costs are 100, and the variable costs per good are Q + 6.
(a) Write down an expression for total revenue, TR, in terms of Q and sketch a graph
of TR against Q, indicating clearly the intercepts with the coordinate axes.
(b) Write down an expression for total costs, TC, in terms of Q and deduce that the
average cost function is given by
AC 5 Q 1 6 1
100
Q
Copy and complete the following table of function values:
Q
2
AC
58
5
10
15
20
Draw an accurate graph of AC against Q and state the value of Q that minimises
average cost.
(c) Show that the profit function is given by
π = 2(2 − Q)(Q − 25)
State the values of Q for which the firm breaks even, and determine the maximum
profit.
150 Chapter 2 Non-linear Equations
Exercise 2.2*
1. If fixed costs are 30, variable costs per unit are Q + 3 and the demand function is
P + 2Q = 50
show that the associated profit function is
π = −3Q2 + 47Q − 30.
Find the break-even values of Q and deduce the maximum profit.
2. The profit function of a firm is of the form
π = aQ2 + bQ + c
If it is known that π = 9, 34 and 19 when Q = 1, 2 and 3, respectively, write down a set
of three simultaneous equations for the three unknowns, a, b and c. Solve this system
to find a, b and c. Hence find the profit when Q = 4.
3. A firm’s average cost function is given by
AC 5
800
1 2Q 1 18
Q
(a) Find, to the nearest whole number, the value of Q at the lowest point on the graph
of AC plotted against Q, in the interval, 0 ≤ Q ≤ 30.
(b) State the value of the fixed costs.
4. If the demand equation is aP + bQ = c, fixed costs are d and variable costs are e per
unit, find expressions, in terms of Q, for each of the following economic functions:
(a) total revenue (b) total cost (c) average cost (d) profit
5. The Ennerdale Bank charges its customers for every withdrawal: $0.50 for each online
transfer and $0.25 for each cash machine withdrawal. The North Borsetshire Bank
charges customers a fixed annual charge of $15 and each debit (online or machine)
costs a further $0.30. You may assume that there are no other withdrawals, that the
account never goes overdrawn and that any interest due is negligible.
(a) The proportion of withdrawals that are via online transfers is a and the total num-
ber of withdrawals made during the year is N. If the cost of operating the two
accounts is the same, show that
a5
1
60
1
5
N
Sketch the graph of this relationship.
(b) What advice can you offer new customers if at least 60% of the customer’s annual
withdrawals are from cash machines?
Section 2.3
Indices and logarithms
Objectives
At the end of this section you should be able to:
●
Evaluate bn in the case when n is positive, negative, a whole number or a fraction.
●
Simplify algebraic expressions using the rules of indices.
●
Investigate the returns to scale of a production function.
●
Evaluate logarithms in simple cases.
●Use
the rules of logarithms to solve equations in which the unknown occurs as a power.
Advice
This section is quite long, with some important ideas. If you are comfortable using the rules of indices
and already know what a logarithm is, you should be able to read through the material in one sitting,
concentrating on the applications. However, if your current understanding is hazy (or non-existent),
you should consider studying this topic on separate occasions. To help with this, the material in this
section has been split into the following convenient sub-sections:
●
●
●
●
index notation;
rules of indices;
logarithms;
summary.
2.3.1 Index notation
We have already used b2 as an abbreviation for b × b. In this section we extend the notation
to bn for any value of n, positive, negative, whole number or fraction. In general, if
M = bn
we say that bn is the exponential form of M to base b. The number n is then referred to as the
index, power or exponent. An obvious way of extending
b2 = b × b
to other positive whole number powers, n, is to define
b3 = b × b × b
b4 = b × b × b × b
and, in general,
bn 5 b 3 b 3 b 3 b 3 . . . 3 b
a total of n
bs multiplied
together
152 Chapter 2 Non-linear Equations
To include the case of negative powers, consider the following table of values of 2n:
2−3
2−2
2−1
20
21
22
23
24
?
?
?
?
2
4
8
16
To work from left to right along the completed part of the table, all you have to do is to
multiply each number by 2. Equivalently, if you work from right to left, you simply divide
by 2. It makes sense to continue this pattern beyond 21 = 2. Dividing this by 2 gives
20 = 2 ÷ 2 = 1
and dividing again by 2 gives
221 5 1 4 2 5 1/2
and so on. The completed table is then
2−3
2−2
2−1
20
21
22
23
24
1
8
1
4
1
2
1
2
4
8
16
Notice that
1
1
5 1
2
2
1
1
222 5 5 2
4
2
1
1
223 5 5 3
8
2
221 5
In other words, negative powers are evaluated by taking the reciprocal of the corresponding
positive power. Motivated by this particular example, we define
b0 5 1
and
1
bn
b2n 5
where n is any positive whole number.
Example
Evaluate
(a) 32
(b) 43
6
(f) (−2)
(g) 3
−4
(c) 70
(h) (−2)
Solution
Using the definitions
bn = b × b × b × . . . × b
b0 = 1
b2n 5
1
bn
(e) 5−1
(d) 51
−3
0
(i) (1.723)
Section 2.3 Indices and logarithms 153
we obtain
(a) 32 = 3 × 3 = 9
(b) 43 = 4 × 4 × 4 = 64
(c) 70 = 1
because any number raised to the power of zero equals 1.
(d) 51 = 5
(e) 521 5
1
1
5
51
5
(f) (−2)6 = (−2) × (−2) × (−2) × (−2) × (−2) × (−2) = 64
where the answer is positive because there are an even number of minus signs.
(g) 324 5
1
1
1
5
5
4
3
3333333
81
(h) (22)23 5
1
1
1
5
52
3
(22) 3 (22) 3 (22)
8
(22)
where the answer is negative because there are an odd number of minus signs.
(i) (1.723)0 = 1
Practice Problem
1. (1) Without using a calculator, evaluate
(a) 102
(b) 101
(c) 100
(d) 10−1
(e) 10−2
(g) (−1)99
(h) 7 −3
(i) (−9)2
( j) (72 101)1
(k) (2.718)0
(f) (−1)100
(2) Confirm your answer to part (1) using a calculator.
We handle fractional powers in two stages. We begin by defining bm where m is a reciprocal such as 1/2 or 1/8 and then consider more general fractions such as 3/4 or 3/8 later. Assuming
that n is a positive whole number, we define
b1/n 5 nth root of b
By this we mean that b1/n is a number which, when raised to the power n, produces b. In
symbols, if c = b1/n then c n = b. Using this definition,
91/2 = square root of 9
= 3 (because 32 = 9)
81/3 = cube root of 8
= 2 (because 23 = 8)
6251/4 = fourth root of 625 = 5 (because 54 = 625)
Of course, the nth root of a number may not exist. There is no number c satisfying c2 = −4, for
example, and so (−4)1/2 is not defined. It is also possible for some numbers to have more than
one nth root. For example, there are two values of c which satisfy c4 = 16, namely c = 2 and
c = −2. In these circumstances it is standard practice to take the positive root, so 161/4 = 2.
154 Chapter 2 Non-linear Equations
We now turn our attention to the case of bm, where m is a general fraction of the form p/q
for some whole numbers p and q. What interpretation are we going to put on a number such
as 163/4? To be consistent with our previous definitions, the numerator, 3, can be thought of
as an instruction for us to raise 16 to the power of 3, and the denominator tells us to take the
fourth root. In fact, it is immaterial in which order these two operations are carried out. If
we begin by cubing 16, we get
163 = 16 × 16 × 16 = 4096
and taking the fourth root of this gives
163/4 = (4096)1/4 = 8 (because 84 = 4096)
On the other hand, taking the fourth root first gives
161/4 = 2 (because 24 = 16)
and cubing this gives
163/4 = 23 = 8
which is the same answer as before. We therefore see that
(163)1/4 = (161/4)3
This result holds for any base b and fraction p/q (provided that q is positive), so we define
b p/q 5 (b p) 1/q 5 (b 1/q) p
To evaluate 25−3/2 it is easier to find the square root first before raising the number to the
power of −3, so
2523/2 5 (251/2)23 5 523 5
1
1
5
3
5
125
For this particular exponential form we have actually carried out three distinct operations.
The minus sign tells us to reciprocate, the fraction, 1/2, tells us to take the square root and the
3 tells us to cube. You might like to check for yourself that you get the same answer irrespective of the order in which these three operations are performed.
Advice
Given that we are allowed to perform these operations in any order, it is usually easier to find the qth
root first to avoid having to spot roots of large numbers.
Practice Problem
2. (1) Without using your calculator, evaluate
(a) 161/2 (b) 271/3 (c) 45/2 (d) 8−2/3 (e) 1−17/25
(2) Confirm your answer to part (1) using a calculator.
SectiOn 2.3 INDICES AND logArITHMS
155
2.3.2 rules of indices
There are two reasons why the exponential form is useful. First, it is a convenient shorthand
for what otherwise might be a very lengthy number. The exponential form
98
is much easier to write down than either of the equivalent forms
9×9×9×9×9×9×9×9
or
43 046 721
Secondly, there are four basic rules of indices which facilitate the manipulation of such
numbers. The four rules may be stated as follows:
Rule 1
bm × bn = bm+n
Rule 2 bm ÷ bn = bm−n
Rule 3 (bm)n = bmn
Rule 4 (ab)n = anbn
It is certainly not our intention to provide mathematical proofs in this text. However, it
might help you to remember these rules if we give you a justification based on some simple
examples. We consider each rule in turn.
rule 1
Suppose we want to multiply together 22 and 25. Now, 22 = 2 × 2 and 25 = 2 × 2 × 2 × 2 × 2, so
22 × 25 = (2 × 2) × (2 × 2 × 2 × 2 × 2)
Notice that we are multiplying together a total of seven 2s and so by definition this is just 27:
that is,
22 × 25 = 27 = 22+5
This confirms rule 1, which tells you that if you multiply two numbers, all you have to do is
to add the indices.
rule 2
Suppose we want to divide 22 by 25. This gives
232
1
1
5
5 3
232323232
23232
2
Now, by definition, reciprocals are denoted by negative indices, so this is just 2−3; that is,
22 ÷ 25 = 2−3 = 22−5
This confirms rule 2, which tells you that if you divide two numbers, all you have to do is to
subtract the indices.
Rule 3
Suppose we want to raise 102 to the power 3. By definition, for any number b,
b3 = b × b × b
156 Chapter 2 Non-linear Equations
so replacing b by 102, we have
(102)3 = 102 × 102 × 102 = (10 × 10) × (10 × 10) × (10 × 10) = 106
because there are six 10s multiplied together: that is,
(102)3 = 106 = 102×3
This confirms rule 3, which tells you that if you take a ‘power of a power’, all you have to do
is to multiply the indices.
Rule 4
Suppose we want to raise 2 × 3 to the power 4. By definition,
b4 = b × b × b × b
so replacing b by 2 × 3 gives
(2 × 3)4 = (2 × 3) × (2 × 3) × (2 × 3) × (2 × 3)
and, because it does not matter in which order numbers are multiplied, this can be written as
(2 × 2 × 2 × 2) × (3 × 3 × 3 × 3)
that is,
(2 × 3)4 = 24 × 34
This confirms rule 4, which tells you that if you take the power of a product of two numbers,
all you have to do is to take the power of each number separately and multiply.
A word of warning is in order regarding these laws. Notice that in rules 1 and 2 the bases
of the numbers involved are the same. These rules do not apply if the bases are different. For
example, rule 1 gives no information about
24 × 35
Similarly, please notice that in rule 4 the numbers a and b are multiplied together. For some
strange reason, some business and economics students seem to think that rule 4 also applies
to addition, so that
(a + b)n = an + bn This statement is NOT true
It would make algebraic manipulation a whole lot easier if it were true, but it is definitely
false! If you need convincing of this, note, for example, that
(1 + 2)3 = 33 = 27
which is not the same as
13 + 23 = 1 + 8 = 9
One variation of rule 4 which is true is
a
b
n
5
an
(b Þ 0)
bn
This is all right because division (unlike addition or subtraction) is the same sort of operation
as multiplication. In fact,
a
b
n
Section 2.3 Indices and logarithms 157
can be thought of as
a3
n
1
b
so applying rule 4 to this product gives
an
1
b
n
5
an
bn
as required.
Advice
There might be occasions (such as in examinations!) when you only half remember a rule or
perhaps think that you have discovered a brand new rule for yourself. If you are ever worried about
whether some rule is legal or not, you should always check it out by trying numbers, just as we did
for (a + b)n. Obviously, one numerical example which actually works does not prove that the rule
will always work. However, one example which fails is good enough to tell you that your supposed
rule is rubbish.
The following example demonstrates how rules 1–4 are used to simplify algebraic
expressions.
Example
Simplify
(a) x1/4 × x3/4 (b)
x2 y3
(c) (x2y−1/3)3
x4 y
Solution
(a) The expression
x1/4 × x3/4
represents the product of two numbers in exponential form with the same base. From
rule 1, we may add the indices to get
x1/4 × x3/4 = x1/4+3/4 = x1
which is just x.
(b) The expression
x2 y3
x4 y
➜
158 Chapter 2 Non-linear Equations
is more complicated than that in part (a) since it involves numbers in exponential form
with two different bases, x and y. From rule 2,
x2
x4
may be simplified by subtracting indices to get
x2 ÷ x4 = x2−4 = x−2
Similarly,
y3
5 y3 4 y1 5 y321 5 y2
y
Hence
x2y3
5 x22y2
x4y
It is not possible to simplify this any further, because x−2 and y2 have different bases.
However, if you prefer, this can be written as
y2
x2
because negative powers denote reciprocals.
(c) An obvious first step in the simplification of
(x2y−1/3)3
is to apply rule 4, treating x2 as the value of a and y−1/3 as b to get
(x2y−1/3)3 = (x2)3(y−1/3)3
Rule 3 then allows us to write
(x2)3 = x2×3 = x6
(y−1/3)3 = y(−1/3)×3 = y−1
Hence
(x2y−1/3)3 = x6y−1
As in part (b), if you think it looks neater, you can write this as
x6
y
because negative powers denote reciprocals.
Practice Problem
3. Simplify
(a) (x3/4)8 (b)
x2
(c) (x2y4)3 (d) x (x 5/2 1 y3)
x 3/ 2
[Hint: in part (d) note that x 5 x1/2 and multiply out the brackets.]
Section 2.3 Indices and logarithms 159
There are occasions throughout this text when we use the rules of indices and definitions
of bn. For the moment, we concentrate on one specific application where we see these ideas in
action. The output, Q, of any production process depends on a variety of inputs, known as
factors of production. These comprise land, capital, labour and enterprise. For simplicity, we
restrict our attention to capital and labour. Capital, K, denotes all man-made aids to production such as buildings, tools and plant machinery. Labour, L, denotes all paid work in the
production process. The dependence of Q on K and L may be written as
Q = f (K, L)
which is called a production function. Once this relationship is made explicit, in the form of a
formula, it is straightforward to calculate the level of production from any given combination of inputs. For example, if
Q = 100K1/3L1/2
then the inputs K = 27 and L = 100 lead to an output
Q = 100(27)1/3(100)1/2
= 100(3)(10)
= 3000
Of particular interest is the effect on output when inputs are scaled in some way. If capital
and labour both double, does the production level also double, does it go up by more than
double or does it go up by less than double? For the particular production function,
Q = 100K1/3L1/2
we see that, when K and L are replaced by 2K and 2L, respectively,
Q = 100(2K)1/3(2L)1/2
Now, by rule 4,
(2K)1/3 = 21/3K1/3 and (2L)1/2 = 21/2L1/2
so
Q = 100(21/3K1/3)(21/2L1/2)
= (21/321/2)(100K1/3L1/2)
The second term, 100K 1/3L1/2, is just the original value of Q, so we see that the output is
multiplied by
21/321/2
Using rule 1, this number may be simplified by adding the indices to get
21/321/2 = 25/6
Moreover, because 5/6 is less than 1, the scale factor is smaller than 2. In fact, my calculator gives
25/6 = 1.78 (to two decimal places)
so output goes up by just less than double.
It is important to notice that the above argument does not depend on the particular value,
2, that is taken as the scale factor. Exactly the same procedure can be applied if the inputs, K
and L, are scaled by a general number l (where l is a Greek letter pronounced ‘lambda’).
Replacing K and L by lK and lL, respectively, in the formula
Q = 100K1/3L1/2
160 Chapter 2 Non-linear Equations
gives
Q = 100(lK)1/3(l L)1/2
(rule 4)
= 100l1/3K1/3l1/2L1/2
= (l l )(100K L )
1/3 1/2
1/3 1/2
(rule 1)
= l5/6(100K1/3L1/2)
We see that the output gets scaled by l5/6, which is smaller than l since the power, 5/6, is less than 1.
We describe this by saying that the production function exhibits decreasing returns to scale.
In general, a function
Q = f (K, L)
is said to be homogeneous if
f (lK, lL) = lnf (K, L)
for some number, n. This means that when both variables K and L are multiplied by l, we can
pull out all of the ls as a common factor, ln. The power, n, is called the degree of homogeneity.
In the previous example we showed that
f (lK, lL) = l5/ 6f (K, L)
and so it is homogeneous of degree 5/6. In general, if the degree of homogeneity, n, satisfies:
●
n < 1, the function is said to display decreasing returns to scale
●
n = 1, the function is said to display constant returns to scale
●
n > 1, the function is said to display increasing returns to scale.
Example
Show that the following production function is homogeneous and find its degree of homogeneity:
Q = 2K1/2L3/2
Does this function exhibit decreasing returns to scale, constant returns to scale or increasing
returns to scale?
Solution
We are given that
f (K, L) = 2K1/2L3/2
so replacing K by lK and L by lL gives
f (lK, l L) = 2(lK )1/2(lL)3/2
We can pull out all of the ls by using rule 4 to get
2l1/2K1/2l3/2L3/2
and then use rule 1 to get
l2(2K1/2L3/2)
l1/2 l3/2 5 l1/213/2
5 l2
Section 2.3 Indices and logarithms 161
We have therefore shown that
f (lK, lL) = l2f (K, L)
and so the function is homogeneous of degree 2. Moreover, since 2 > 1, we deduce that it
has increasing returns to scale.
Practice Problem
4. Show that the following production functions are homogeneous and comment on their
returns to scale:
(a) Q = 7KL2 (b) Q = 50K1/4L3/4
You may well have noticed that all of the production functions considered so far are of the
form
Q = AK aLb
for some positive constants, A, a and b. (The Greek letters a and b are pronounced ‘alpha’
and ‘beta’, respectively.) Such functions are called Cobb–Douglas production functions. It is
easy to see that they are homogeneous of degree a + b because if
f (K, L) = AKaLb
then
f (lK, lL) = A(l K)a(lL)b
= AlaK albLb
(rule 4)
= la+b(AKaLb)
(rule 1)
=l
f (K, L)
a+b
Consequently, Cobb–Douglas production functions exhibit
●
decreasing returns to scale, if a + b < 1
●
constant returns to scale, if a + b = 1
●
increasing returns to scale, if a + b > 1.
By the way, not all production functions are of this type. Indeed, it is not even necessary for
a production function to be homogeneous. Some examples illustrating these cases are given in
Question 5 in Exercise 2.3 at the end of this section. We shall return to the topic of production
functions in Chapter 5.
2.3.3 Logarithms
At the beginning of this section we stated that if a number, M, is expressed as
M = bn
162
chapter 2 NoN-lINEAr EQuATIoNS
then bn is called the exponential form of M to base b. The approach taken so far has simply
been to evaluate M from any given values of b and n. In practice, it may be necessary to
reverse this process and to find n from known values of M and b. To solve the equation
32 = 2n
we need to express 32 as a power of 2. In this case it is easy to work out n by inspection.
Simple trial and error easily gives n = 5 because
25 = 32
We describe this expression by saying that the logarithm of 32 to base 2 is 5. In symbols
we write
log232 = 5
Quite generally,
if M = bn then
logbM = n
where n is called the logarithm of M to base b.
advice
Students have been known to regard logarithms as something rather abstract and difficult to understand. There is, however, no need to worry about logarithms, since they simply provide an alternative
way of thinking about numbers such as bn. read through the following example and then try Practice
Problem 5 for yourself. You might discover that they are easier than you expect.
example
Evaluate
(a) log39
(b) log42
(c) log71/7
Solution
(a) To find the value of log39 we convert the problem into one involving powers. From the
definition of a logarithm to base 3, we see that the statement
log39 = n
is equivalent to
9 = 3n
The problem of finding the logarithm of 9 to base 3 is exactly the same as that of writing 9 as a power of 3. The solution of this equation is clearly n = 2 since
9 = 32
Hence log39 = 2.
SectiOn 2.3 INDICES AND logArITHMS
163
(b) Again, to evaluate log42 we merely rewrite
log42 = n
in exponential form as
2 = 4n
The problem of finding the logarithm of 2 to base 4 is exactly the same as that of writing
2 as a power of 4. The value of 2 is obtained from 4 by taking the square root, which
involves raising 4 to the power of 1/2, so
2 = 41/2
Hence log42 = 1/2.
(c) If
log71/7 = n
then
/7 = 7n
1
The value of 1/7 is found by taking the reciprocal of 7, which involves raising 7 to the
power of −1: that is,
/7 = 7−1
1
Hence log71/7 = −1.
practice problem
5. (1) Write down the values of n which satisfy
(a) 1000 = 10n
(b) 100 = 10n
(d) 1 = 10n
(e)
1
5 10n
10
(c) 10 = 10n
(f)
1
5 10n
100
(2) Use your answer to part (1) to write down the values of
(a) log101000
(d) log101
(b) log10100
1
(e) log10 /10
(c) log1010
(f) log101/100
(3) Confirm your answer to part (2) using a calculator.
Given the intimate relationship between exponentials and logarithms, you should not be
too surprised to learn that logarithms satisfy three rules that are comparable with those for
indices. The rules of logarithms are as follows:
Rule 1 logb(x × y) = logbx + logby
Rule 2 logb(x ÷ y) = logbx − logby
Rule 3 logbxm = m logbx
The following two examples show how to use these rules to simplify algebraic expressions
and how to solve equations in which the unknown appears as a power.
164 Chapter 2 Non-linear Equations
Example
Use the rules of logarithms to express each of the following as a single logarithm:
(a) logb x + logby − logbz (b) 2logb x − 3logby
Solution
(a) The first rule of logs shows that the sum of two logs can be written as the log of a
product, so
logbx + logby − logbz = logb(xy) − logbz
Also, according to rule 2, the difference of two logs is the log of a quotient, so we can
simplify further to get
log b
xy
z
(b) Given any combination of logs such as
2logbx − 3logby
the trick is to use the third rule to ‘get rid’ of the coefficients. Since
2logbx = logbx2
and
3logby = logby3
we see that
2logbx − 3logby = logbx2 − logby3
Only now can we use the second rule of logs, which allows us to write the expression
as the single logarithm
log b
x2
y3
Practice Problem
6. Use the rules of logs to express each of the following as a single logarithm:
(a) log b x − logb y + logbz (b) 4logb x + 2logby
Before we continue with this topic, a word of warning is in order. Be careful to learn the
rules of logs correctly. A common mistake is to misread rule 1 as
logb(x + y) = logb x + logby This is NOT true
Remember that logs are just a posh way of thinking about indices and it is when you multiply
numbers together that you end up adding the indices, so the correct version has to be
logb(xy) = logb x + logby
Section 2.3 Indices and logarithms 165
Example
Find the value of x which satisfies
(a) 200(1.1)x = 20 000 (b) 5x = 2(3)x
Solution
(a) An obvious first step in the solution of
200(1.1)x = 20 000
is to divide both sides by 200 to get
(1.1)x = 100
In Chapter 1 it was pointed out that we can do whatever we like to an equation, provided that we do the same thing to both sides. In particular, we may take logarithms
of both sides to get
log(1.1)x = log(100)
Now by rule 3 we have
log(1.1)x = x log(1.1)
so the equation becomes
x log(1.1) = log(100)
Notice the effect that rule 3 has on the equation. It brings the unknown down to the
same level as the rest of the expression. This is the whole point of taking logarithms,
since it converts an equation in which the unknown appears as a power into one which
can be solved using familiar algebraic methods. Dividing both sides of the equation
x log(1.1) = log(100)
by log(1.1) gives
x5
log(100)
log(1.1)
So far no mention has been made of the base of the logarithm. The above equation for
x is true no matter what base is used. It makes sense to use logarithms to base 10
because all scientific calculators have this facility as one of their function keys. Using
base 10, my calculator gives
x5
log(100)
2
5
5 48.32
log(1.1) 0.041 395 685
check this using
your own calculator
to two decimal places.
As a check, if this number is substituted back into the original equation, then
200(1.1)x = 200(1.1)48.32 = 20 004 ✓
We cannot expect to obtain the exact answer, because we rounded x to only 2 decimal
places.
➜
166 Chapter 2 Non-linear Equations
(b) To solve
5x = 2(3)x
we take logarithms of both sides to get
log(5x) = log(2 × 3x)
The right-hand side is the logarithm of a product and, according to rule 1, can be
written as the sum of the logarithms, so the equation becomes
log(5x) = log(2) + log(3x)
As in part (a), the key step is to use rule 3 to ‘bring down the powers’. If rule 3 is
applied to both log(5x) and log(3x), then the equation becomes
x log(5) = log(2) + x log(3)
This is now the type of equation that we know how to solve. We collect xs on the
left-hand side to get
x log(5) − x log(3) = log(2)
and then pull out a common factor of x to get
x[log(5) − log(3)] = log(2)
Now, by rule 2, the difference of two logarithms is the same as the logarithm of their
quotient, so
log(5) − log(3) = log(5 ÷ 3)
Hence the equation becomes
5
5 log(2)
3
x log
so
x5
log(2)
log(5/3)
Finally, taking logarithms to base 10 using a calculator gives
x5
0.301 029 996
5 1.36
0.221 848 750
to two decimal places.
As a check, the original equation
5x = 2(3)x
becomes
51.36 = 2(3)1.36
that is,
8.92 = 8.91 ✓
Again, the slight discrepancy is due to rounding errors in the value of x.
Section 2.3 Indices and logarithms 167
Practice Problem
7. Solve the following equations for x:
(a) 3x = 7 (b) 5(2)x = 10x
Advice
In this section we have met a large number of definitions and rules concerning indices and
logarithms. For convenience, we have collected these together in the form of a summary. The facts
relating to indices are particularly important, and you should make every effort to mem­orise these
before proceeding with the rest of this text.
2.3.4 Summary
Indices
If n is a positive whole number, then
bn = b × b × . . . × b
b0 = 1
b−n = 1/bn
b1/n = nth root of b
Also, if p and q are whole numbers with q > 0, then
b p/q = (b p)1/q = (b1/q) p
The four rules of indices are:
Rule 1
bm × bn = bm+n
Rule 2 bm ÷ bn = bm−n
Rule 3 (bm)n = bmn
Rule 4 (ab)n = anbn
Logarithms
If M = bn, then n = logb M. The three rules of logarithms are:
Rule 1
logb(x × y) = logbx + logby
Rule 2
logb(x ÷ y) = logbx − logby
Rule 3 logbxm = m logbx
168 Chapter 2 Non-linear Equations
Key Terms
Capital Man-made assets used in the production of goods and services.
Cobb–Douglas production function
A production function of the form: Q = AK aLb.
Constant returns to scale Exhibited by a production function when a given percentage increase
in input leads to the same percentage increase in output: f(lK, lL) = lf(K, L).
Decreasing returns to scale Exhibited by a production function when a given percentage increase
in input leads to a smaller percentage increase in output: f(lK, lL) = lnf(K, L) where 0 < n < 1.
Degree of homogeneity
The number n in the relation f(lK, lL) = lnf(K, L).
Exponent A superscript attached to a variable; the number 5 is the exponent in the expression, 2x5.
Exponential form A representation of a number which is written using powers. For example, 25 is
the exponential form of the number 32.
Factors of production The inputs into the production of goods and services: labour, land, capital
and raw materials.
Homogeneous function A function with the property that when all of the inputs are multiplied by
a constant, l, the output is multiplied by ln where n is the degree of homogeneity.
Increasing returns to scale Exhibited by a production function when a given percentage increase
in input leads to a larger percentage increase in output: f(lK, lL) = lnf(K, L) where n > 1.
Index Alternative word for exponent or power.
Labour All forms of human input to the production process.
Logarithm The power to which a base must be raised to yield a particular number.
Power Another word for exponent. If this is a positive integer, then it gives the number of times a
number is multiplied by itself.
Production function The relationship between the output of a good and the inputs used to produce it.
Exercise 2.3
1. (1) Without using your calculator, evaluate
(a) 82
(b) 21
(c) 3−1
(d) 170
(e) 11/5
(f) 361/2
(g) 82/3
(h) 49−3/2
(2) Confirm your answer to part (1) using a calculator.
2. Use the rules of indices to simplify
b7
b2
(a) a3 × a8
(b)
(f) y3 ÷ y7
(g) (x1/2)8
x4 y5
x2 y3
(c) (c2)3
(d)
(h) f 2 × f 4 × f
(i) ( y 6 )
(e) (xy2)3
( j)
x3
x 22
3. Write the following expressions using index notation:
(a) x (b)
1
1
1
(c) 3 x (d) (e)
(f) x x
2
x
x
x
4. For the production function, Q = 200K1/4L2/3, find the output when
(a) K = 16, L = 27 (b) K = 10 000, L = 1000
5. Which of the following production functions are homogeneous? For those functions
which are homogeneous, write down their degrees of homogeneity and comment on
their returns to scale.
Section 2.3 Indices and logarithms 169
(a) Q = 500K1/3L1/4
(b) Q = 3LK + L2
(c) Q = L + 5L2K 3
6. Write down the values of x which satisfy each of the following equations:
1
3
(a) 5x = 25
(b) 3x 5
(d) 2x = 64
(e) 100x = 10
(c) 2x 5
1
8
(f) 8x = 1
7. Write down the value of
(a) logbb2
(b) logbb
(c) logb1
(d) log b b (e) logb(1/b)
8. Use the rules of logs to express each of the following as a single log:
(a) logb x + logbz
(b) 3logb x − 2logby
(c) logb y − 3logb z
9. Express the following in terms of logb x and logby:
(a) logb x 2y
(b) log b
x
y2
(c) logb x 2y7
10. Solve the following equations for x. Give your answers to two decimal places.
(a) 5x = 8 (b) 10x = 50 (c) 1.2x = 3 (d) 1000 × 1.05x = 1500
11. (1) State the values of
(a) log232 (b) log 9
1
3
(2) Use the rules of logs to express
2logb x − 4logby
as a single logarithm.
(3) Use logs to solve the equation
10(1.05)x = 300
Give your answer correct to one decimal place.
12. (1) State the values of x that satisfy the following equations:
(a) 81 = 3x (b)
1
5 5x (c) 161/2 = 2x
25
(2) Use the rules of indices to simplify:
(a)
x6 y9
x 3 y8
(b) (x3y)5
(c)
x9y4
x5
➜
170 Chapter 2 Non-linear Equations
13. The number of complaints, N, received by a small company each month can be
modelled by
N = 80log10(7 + 10t)
where t denotes the number of months since the company’s launch.
(a) Estimate the number of complaints received by the company each month for the
first six months of trading.
(b) Plot a graph of N against t and hence comment on how N varies with t.
14. If two firms A and B use the same labour input, L, their output in the short term is given
by QA = 108 L and QB = 4L2, respectively. Find the non-zero value of L which produces
the same level of output for these two firms.
Exercise 2.3*
1. (1) Evaluate the following without using a calculator:
(a) 323/5 (b) 64−5/6 (c)
1
125
24/3
(d) 3
3
8
2/ 3
(e) 2
(2) Confirm your answer to part (1) using a calculator.
2. Use the rules of indices to simplify
(a) y 3/2 × y1/2
(b)
x2y
xy]1
(c) (xy1/2)4
(d) (p2)1/3 ÷ (p1/3)2 (e) (24q)1/3 ÷ (3q)1/3 (f) (25p2q4)1/2
3. Write the following expressions using index notation:
(a)
8
1
1
(b) 4 x (c)
(d) 2 x 5 x (e) 3
x7
x( x )
x x
4. If a 5
2 x
4b
and b = 3x4y, simplify 2 .
y3
a
5. Show that the production function
Q = A[bK a + (1 − b)La]1/a
is homogeneous and displays constant returns to scale.
6. Solve the following equations:
1
(a) 2 = 4 (b) 4 × 2 = 32 (c) 8 5 2 3
2
3x
x
x
x
7. Use the rules of logs to express each of the following as a single log:
(a) logb(xy) − logb x − logby
(b) 3logb x − 2logby
(c) logby + 5logb x − 2logb z
(d) 2 + 3logb x
1
4
]1/ 2
Section 2.3 Indices and logarithms 171
8. Express the following in terms of logb x, logby and logb z:
(a) logb(x 2y3z4) (b) logb
x4
(c) logb
y2z5
x
yz
9. If logb2 = p, logb3 = q and logb10 = r, express the following in terms of p, q and r:
(a) logb
1
3
(b) logb12
(c) logb0.000 3
(d) logb600
10. Solve the following equations. Round your answers to two decimal places.
(a) 10(1.07)x = 2000
(b) 10x−1 = 3
(c) 5x−2 = 5
(d) 2(7)−x = 3x
11. Solve the inequalities giving the bounds to three decimal places:
(a) 32x+1 ≤ 7
(b) 0.8x < 0.04
12. Solve the equation
log10(x 1 2) 1 1og10 x 2 l 5 1og10
3
2
13. (1) Define the term homogeneous when used to describe a production function f (K, L).
(2) If the production function
f (K, L) = 4K mL1/3 + 3K
is homogeneous, state the value of m.
Does the function display decreasing, constant or increasing returns to scale?
14. (1) State the values of x that satisfy the following equations:
(a) 4 = 8x (b) 5 5
1
25
x
(2) Express y in terms of x:
2loga x = loga7 + logay
15. Show that 2log10 x 2
log10
x4
y
1
1
log10 y 2 log10 1000 can be simplified to
2
3
21
16. Transpose each of the following production functions for L:
(a) Q = AKaLb (b) Q = A[bKa + (1 − b)La ]1/a
17. Show that each of these functions is homogeneous and state the degree of homogeneity:
(a) f (K,L) 5
K 2 1 L2
K1L
K 2 1 L2
KL
(c) f (K,L) = A[aKm + bLm]n/m
(b) f (K,L) 5 KL log
(d) f (K,L) = KL2g(L/K) where g is a general function.
Section 2.4
The exponential and natural
logarithm functions
Objectives
At the end of this section you should be able to:
Sketch graphs of general exponential functions.
●
●Understand
how the number e is defined.
●Use
the exponential function to model growth and decay.
●Use
log graphs to find unknown parameters in simple models.
●Use
the natural logarithm function to solve equations.
In the previous section we described how to define numbers of the form bx, and we discussed
the idea of a logarithm, logbx. It turns out that there is one base (the number e = 2.718 281 . . .)
that is particularly important in mathematics. The purpose of this section is to introduce you
to this strange number and to consider a few simple applications.
We begin by investigating the graphs of the functions,
f (x) = 2x and
g(x) = 2−x
As we pointed out in Section 2.3, a number such as 2x is said to be in exponential form. The
number 2 is called the base and x is called the exponent. Values of this function are easily
found either by pressing the power key x y on a calculator or by using the definition of b n given
in Section 2.3. A selection of these is given in the following table:
x
2
x
−3
−2
−1
0
1
2
3
4
5
0.125
0.25
0.5
1
2
4
8
16
32
A graph of f (x) based on this table is sketched in Figure 2.12. Notice that the graph
approaches the x axis for large negative values of x, and it rises rapidly as x increases.
A graph of the negative exponential, g(x) = 2−x, shown in Figure 2.13, is based on the
following table of values:
x
2−x
−5
−4
−3
−2
−1
0
1
2
3
32
16
8
4
2
1
0.5
0.25
0.125
This function is sketched in Figure 2.13. It is worth noticing that the numbers appearing
in the table of 2−x are the same as those of 2x but arranged in reverse order. Hence the graph
of 2−x is obtained by reflecting the graph of 2x in the y axis.
Figure 2.12 displays the graph of a particular exponential function, 2x. Quite generally, the
graph of any exponential function
f (x) = bx
Section 2.4 The exponential and natural logarithm functions 173
y
36
32
28
24
y 5 2x
20
16
12
8
4
24 23 22 21
1
2
3
4
5
x
Figure 2.12
y
36
32
28
24
y52
2x
20
16
12
8
4
25 24 23 22 21
1
2
3
4
5
x
Figure 2.13
has the same basic shape provided b > 1. The only difference is that larger values of b produce steeper curves. A similar comment applies to the negative exponential, b−x.
Practice Problem
1. Complete the following table of function values of 3x and 3−x and hence sketch their graphs.
x
x
3
3−x
−3
−2
−1
0
1
2
3
174 Chapter 2 Non-linear Equations
Obviously, there is a whole class of functions, each corresponding to a different base, b. Of
particular interest is the case when b takes the value
2.718 281 828 459 . . .
This number is written as e and the function
f (x) = ex
is referred to as THE exponential function. In fact, it is not necessary for you to understand
where this number comes from. All scientific calculators have an ex button, and you may
simply wish to accept the results of using it. However, it might help your confidence if you
have some appreciation of how it is defined. To this end, consider the following example and
subsequent problem.
Example
Evaluate the expression
11
m
1
m
where m = 1, 10, 100 and 1000, and comment briefly on the behaviour of this sequence.
Solution
Substituting the values m = 1, 10, 100 and 1000 into
11
m
1
m
gives
1
11
1
11
1
1
10
5 21 5 2
10
1
11
100
11
1
1000
5 (1.1)10 5 2.593 742 460
100
5 (1.01)100 5 2.704 813 829
1000
5 (1.001)1000 5 2.716 923 932
The numbers are clearly getting bigger as m increases. However, the rate of increase
appears to be slowing down, suggesting that numbers are converging to some fixed value.
The following problem gives you an opportunity to continue the sequence and to discover
for yourself the limiting value.
Section 2.4 The exponential and natural logarithm functions 175
Practice Problem
2. (a) Use the power key xy on your calculator to evaluate
1
11
m
m
where m = 10 000, 100 000 and 1 000 000.
(b) Use your calculator to evaluate e1 and compare with your answer to part (a).
Hopefully, the results of Practice Problem 2 should convince you that as m gets larger, the
value of
1
11
m
m
approaches a limiting value of 2.718 281 828 . . . , which we choose to denote by the letter e.
In symbols we write
e 5 lim 1 1
m
`
1
m
m
The significance of this number can only be fully appreciated in the context of calculus,
which we study in Chapter 4. However, it is useful at this stage to consider some preliminary
examples. These will give you practice in using the ex button on your calculator and will give
you some idea how this function can be used in modelling.
Advice
The number e has a similar status in mathematics as the number π and is just as useful. It arises in
the mathematics of finance, which we discuss in the next chapter. You might like to glance through
Section 3.2 now if you need convincing of the usefulness of e.
Example
The percentage, y, of households possessing refrigerators, t years after they have been
introduced in a developed country, is modelled by
y = 100 − 95e−0.15t
(1) Find the percentage of households that have refrigerators
(a) at their launch;
(b) after 1 year;
(c) after 10 years;
(d) after 20 years.
➜
176 Chapter 2 Non-linear Equations
(2) What is the market saturation level?
(3) Sketch a graph of y against t and hence give a qualitative description of the growth of
refrigerator ownership over time.
Solution
(1) To calculate the percentage of households possessing refrigerators now and in 1, 10
and 20 years’ time, we substitute t = 0, 1, 10 and 20 into the formula
y = 100 − 95e−0.15t
to get
(a) y(0) = 100 − 95e0 = 5%
(b) y(1) = 100 − 95e−0.15 = 18%
check these numbers
on your own calculator
(c) y(10) = 100 − 95e−1.5 = 79%
(d) y(20) = 100 − 95e−3.0 = 95%
(2) To find the saturation level, we need to investigate what happens to y as t gets ever
larger. We know that the graph of a negative exponential function has the basic shape
shown in Figure 2.13. Consequently, the value of e−0.15t will eventually approach zero
as t increases. The market saturation level is therefore given by
y = 100 − 95(0) = 100%
(3) A graph of y against t, based on the information obtained in parts (1) and (2), is
sketched in Figure 2.14.
This shows that y grows rapidly to begin with, but slows down as the market
approaches saturation level. An economic variable which increases over time but
approaches a fixed value like this is said to display limited growth. A saturation level of
100% indicates that eventually all households are expected to possess refrigerators,
which is not surprising given the nature of the product.
y
100
90
80
70
60
50
40
y 5 100 2 95e20.15t
30
20
10
2
Figure 2.14
4
6
8
10
12
14
16
18
20
22
24
t
SectiOn 2.4 THE ExPoNENTIAl AND NATurAl logArITHM FuNCTIoNS
177
practice problem
3. The percentage, y, of households possessing microwave ovens t years after they have
been launched is modelled by
y5
55
1 1 800e20.3t
(1) Find the percentage of households that have microwaves
(a) at their launch;
(b) after 10 years;
(c) after 20 years;
(d) after 30 years.
(2) What is the market saturation level?
(3) Sketch a graph of y against t and hence give a qualitative description of the growth
of microwave ownership over time.
In Section 2.3 we noted that if a number M can be expressed as bn, then n is called the
logarithm of M to base b. In particular, for base e,
if M = en then n = logeM
We call logarithms to base e natural logarithms. These occur sufficiently frequently to warrant
their own notation. Rather than write logeM, we simply put ln M instead. The three rules of
logs can then be stated as
Rule 1 ln(x × y) = ln x + ln y
Rule 2 ln(x ÷ y) = ln x − ln y
Rule 3 ln xm = m ln x
example
Use the rules of logs to express
(a) ln
x
in terms of ln x and ln y
y
(b) 3 ln p + ln q − 2 ln r as a single logarithm.
Solution
(a) In this part we need to ‘expand’, so we read the rules of logs from left to right:
ln
x
y
5 ln x 2 ln y
(rule 2)
5 ln x 2 ln y1/2
(fractional powers denote roots)
1
5 ln x 2 ln y
2
(rule 3 )
➜
178 Chapter 2 Non-linear Equations
(b) In this part we need to reverse this process and so read the rules from right to left:
3ln p 1 ln q 2 2ln r 5 ln p3 1 ln q 2 ln r 2
3
5 ln (p q) 2 ln r
2
3
5 ln
pq
r2
(rule 3)
(rule 1)
(rule 2)
Practice Problem
4. Use the rules of logs to express
(a) ln(a2b3) in terms of ln a and ln b
(b) 1./2 ln x − 3 ln y as a single logarithm.
As we pointed out in Section 2.3, logs are particularly useful for solving equations in which
the unknown occurs as a power. If the base is the number e, then the equation can be solved
by using natural logarithms.
Example
An economy is forecast to grow continuously so that the gross national product (GNP),
measured in billions of dollars, after t years is given by
GNP = 80e0.02t
After how many years is GNP forecast to be $88 billion? What does the model predict
about the value of GNP in the long run?
Solution
We need to solve
88 = 80e0.02t
for t. Dividing through by 80 gives
1.1 = e0.02t
Using the definition of natural logarithms we know that
if M = en then n = ln M
Section 2.4 The exponential and natural logarithm functions 179
If we apply this definition to the equation
1.1 = e0.02t
we deduce that
0.02t = ln 1.1 = 0.095 31 . . . (check this using your own calculator)
so
t5
0.095 31
5 4.77
0.02
We therefore deduce that GNP reaches a level of $88 billion after 4.77 years.
A graph of GNP plotted against time would be similar in shape to the graph in
Figure 2.14. This shows that GNP just keeps on rising over time (in fact, at an increasing
rate). Such a model is said to display unlimited growth.
Practice Problem
5. During a recession, a firm’s revenue declines continuously so that the revenue, TR
(measured in millions of dollars), in t years’ time is modelled by
TR = 5e−0.15t
(a) Calculate the current revenue and also the revenue in two years’ time.
(b) After how many years will the revenue decline to $2.7 million?
One important (but rather difficult) problem in modelling is to extract a mathematical
formula from a table of numbers. If this relationship is of the form of an exponential, then it
is possible to estimate values for some of the parameters involved.
Advice
The following example shows how to find such a formula from data points. This is an important
skill. However, it is not crucial to your understanding of subsequent material in this text. You
may wish to miss this out on first reading and move straight on to the exercises at the end of this
chapter.
180 Chapter 2 Non-linear Equations
Example
The values of GNP, g, measured in billions of dollars, over a period of t years was
observed to be
t (years)
g (billions of dollars)
2
12
5
16
10
27
20
74
Model the growth of GNP using a formula of the form
g = BeAt
for appropriate values of A and B. Hence estimate the value of GNP after 15 years.
Solution
Figure 2.15 shows the four points plotted with g on the vertical axis and t on the horizontal
axis. The basic shape of the curve joining these points certainly suggests that an exponential
function is likely to provide a reasonable model, but it gives no information about what
values to use for the parameters A and B. However, since one of the unknown parameters,
A, occurs as a power in the relation
g = BeAt
it is a good idea to take natural logs of both sides to get
ln g = ln(BeAt )
The rules of logs enable us to expand the right-hand side to get
ln(BeAt ) = ln B + ln(eAt ) (rule 1)
(definition of a log to base e)
= ln B + At
Hence
ln g = At + ln B
g
100
90
80
70
60
50
40
30
20
10
2
Figure 2.15
4
6
8
10
12
14
16
18
20
22
t
Section 2.4 The exponential and natural logarithm functions 181
Although this does not look like it at first sight, this relation is actually the equation of
a straight line! To see this, recall that the usual equation of a line is y = ax + b. The log
equation is indeed of this form if we put
y = ln g and x = t
The equation then becomes
y = Ax + ln B
so a graph of ln g plotted on the vertical axis with t plotted on the horizontal axis should
produce a straight line with slope A and with an intercept on the vertical axis of ln B.
Figure 2.16 shows this graph based on the table of values
x=t
2
5
10
20
y = ln g
2.48
2.77
3.30
4.30
As one might expect, the points do not exactly lie on a straight line, since the formula is
only a model. However, the line sketched in Figure 2.16 is a remarkably good fit. The slope
can be calculated as
A5
423
5 0.09
18.6 2 7.6
and the vertical intercept can be read off the graph as 2.25. This is ln B and so
B = e2.25 = 9.49
Hence the formula for the approximate relation between g and t is
g = 9.49e0.09t
An estimate of the GNP after 15 years can be obtained by substituting t = 15 into this
formula to get
g = 36.6 (billion dollars)
In g
5
4
1
3
2.25
11
2
1
2
Figure 2.16
4
6
8
10
12
14
16
18
20
t
182 Chapter 2 Non-linear Equations
Practice Problem
6. Immediately after the launch of a new product, the monthly sales figures (in thousands)
are as follows:
t (months)
1
3
6
12
s (sales)
1.8
2.7
5.0
16.5
(1) Complete the following table of values of ln s:
t
1
ln s
0.59
3
6
12
1.61
(2) Plot these points on graph paper with the values of ln s on the vertical axis and t on
the horizontal axis. Draw a straight line passing close to these points. Write down
the value of the vertical intercept and calculate the slope.
(3) Use your answers to part (2) to estimate the values of A and B in the relation
s = BeAt.
(4) Use the exponential model derived in part (3) to estimate the sales when
(a) t = 9 (b) t = 60
Which of these estimates would you expect to be the more reliable? Give a reason
for your answer.
Key Terms
Exponential function The function f(x) = ex; an exponential function in which the base is the
number e = 2.718 281. . . . Limited growth Used to describe an economic variable which increases over time but which tends
to a fixed quantity.
Natural logarithm
write, n = ln M.
A logarithm to base e; if M = en then n is the natural logarithm of M and we
Unlimited growth
Used to describe an economic variable which increases without bound.
Exercise 2.4
1. The number of items, N, produced each day by an assembly-line worker, t days after
an initial training period, is modelled by
N = 100 − 100e−0.4t
(1) Calculate the number of items produced daily
(a) 1 day after the training period;
(b) 2 days after the training period;
(c) 10 days after the training period.
Section 2.4 The exponential and natural logarithm functions 183
(2) What is the worker’s daily production in the long run?
(3) Sketch a graph of N against t and explain why the general shape might have been
expected.
2. Use the rules of logs to expand each of the following:
(a) ln xy
(d) ln
x5
y7
(b) ln xy4
(e) ln
x
y
(c) ln (xy)2
(f) ln
xy 3
z
3. Use the rules of logs to express each of the following as a single logarithm:
(a) ln x + 2 ln x
(b) 4 ln x − 3 ln y + 5 ln z
4. Solve each of the following equations. (Round your answer to two decimal places.)
(a) ex = 5.9
(b) ex = 0.45
(c) ex = −2
(d) e3x = 13.68
(e) e−5x = 0.34
(f) 4e2x = 7.98
5. The value of a second-hand car reduces exponentially with age, so that its value $y after
t years can be modelled by the formula
y = Ae−ax
If the car was worth $50 000 when new and $38 000 after two years, find the values of
A and a, correct to three decimal places.
Use this model to predict the value of the car
(a) when the car is five years old;
(b) in the long run.
6. Solve the following equations:
(a) ln x = 5
(b) ln x = 0
7. Future sales of two products A and B are given by SA = 5e0.01t and SB = 2e0.02t. Find the
time, t, when sales of the two products are the same.
8. Show that the following production function is homogeneous and state whether it
displays decreasing, increasing or constant returns to scale.
f (K,L) = (K 2 + L2)eK/L
Exercise 2.4*
1. The value (in cents) of shares, t years after their flotation on the stock market, is modelled by
V = 6e0.8t
Find the increase in the value of these shares, four years and two months later. Give
your answer to the nearest cent.
2. Solve each of the following equations, correct to two decimal places:
(a) 6e−2x = 0.62
(b) 5 ln(4x) = 9.84
(c) 3 ln(5x) − 2 ln(x) = 7
➜
184 Chapter 2 Non-linear Equations
3. A team of financial advisers guiding the launch of a national newspaper has modelled
the future circulation of the newspaper by the equation
N = c(1 − e−kt )
where N is the daily circulation after t days of publication, and c and k are positive
constants. Transpose this formula to show that
t5
1
c
ln
k
c]N
When the paper is launched, audits show that
and k = 1 ln 2
30
(a) Calculate the daily circulation after 30 days of publication.
c = 700 000
(b) After how many days will the daily circulation first reach 525 000?
(c) What advice can you give the newspaper proprietor if it is known that the paper will
break even only if the daily circulation exceeds 750 000?
4. A Cobb–Douglas production function is given by
Q = 3L1/2K1/3
Find an expression for ln Q in terms of ln L and ln K.
If a graph were to be sketched of ln Q against ln K (for varying values of Q and K
but with L fixed), explain briefly why the graph will be a straight line and state its slope
and vertical intercept.
5. The following table gives data relating a firm’s output, Q, and labour, L:
L
1
2
3
4
5
Q
0.50
0.63
0.72
0.80
0.85
The firm’s short-run production function is believed to be of the form
Q = ALn
(a) Show that
ln Q = n ln L + ln A
(b) Using the data supplied, copy and complete the following table:
ln L
ln Q
0.69
−0.69
1.39
−0.33
−0.16
Plot these points with ln L on the horizontal axis and ln Q on the vertical axis. Draw
a straight line passing as close as possible to all five points.
(c) By finding the slope and vertical intercept of the line sketched in part (b), estimate
the values of the parameters n and A.
6. (a) Multiply out the brackets
(3y − 2)(y + 5)
Section 2.4 The exponential and natural logarithm functions 185
(b) Solve the equation
3e2x + 13ex = 10
Give your answer correct to three decimal places.
7. (a) Make y the subject of the equation
x = aeby
(b) Make x the subject of the equation
y = ln(3 + e2x )
8. Solve the following equations for x:
(a) ln(x − 5) = 0
(d) e5x+1 = 10
(b) ln(x2 − x − 1) = 0
−x2/2
(e) e
(c) x ln( x − 4) = 0
= 0.25
9. The demand and supply functions of a good are given by
QD = Ae−k1P and
QS = Bek2P respectively
where A, B, k1 and k2 are positive constants.
Find the equilibrium price and show that the equilibrium quantity is given by
1
(Ak2Bk1) k21k1 .
Formal mathematics
The quadratic functions that we have investigated in this chapter have an obvious extension
to cubic functions:
f(x) = ax3 + bx2 + cx + d
Linear, quadratic and cubic functions are all examples of a general class of functions called
polynomials which are defined by
f(x) = an x n + an−1x n−1 + . . . + a0
The coefficients, ai, are constants, and the highest power of x is called the degree of the polynomial. A quadratic is a polynomial of degree 2, and a cubic has degree 3. Functions such as
f (x) 5
1
1
, however, are not polynomials.
1 4 and f (x) 5 2
x
x 2 2x 1 1
One important property of functions is that of continuity. A simple way of understanding
this is to imagine drawing the graph of a function. If this can be done without taking your
pen off the page, the function is described as being continuous everywhere. An example of
such a function is shown in Figure 2.17(a). Polynomials are examples of functions which are
continuous at all values of their domain. On the other hand, if the graph has jumps or breaks
in it, then the function is discontinuous at those points. Figure 2.17(b) shows the graph of a
function which is undefined at x = 2, and the graph separates into two different branches on
either side of x = 2. Functions involving reciprocals are examples of functions of this type.
1
. The graph
x22
in Figure 2.17(c) also illustrates a function which is discontinuous at x = 2 because it has a
Indeed, the graph in Figure 2.17(a) is that of the specific function, f (x) 5
jump at this point. For values of x < 2, the function takes the constant value of 3, whereas
when x ≥ 2, the function takes the constant value of 5. In other words,
f (x) 5
3, x , 2
5, x $ 2
A function which is defined in separate bits like this is said to be defined piecewise.
5
3
2
(a)
Figure 2.17
(b)
2
(c)
Formal mathematics 187
The majority of functions in economics are continuous, reflecting the fact that usually
there are gradual fluctuations in economic variables without any sudden changes. However,
sometimes there are catastrophic events which lead to sudden jumps. For example, the
outbreak of war (or rumour of one) can lead to dramatic changes in prices of commodities
such as oil or gold.
The above description of continuous functions is easy to visualise but is not very precise.
In order to give a formal definition of a continuous function, we need the concept of a limit. We
write
(read ‘the limit of f of x, as x tends to a’)
lim f (x)
x
a
for the value that the function gets ever closer to, as x approaches a value a. This is most
easily understood via an example.
Example
ex 2 1
0
x
Find lim
x
Solution
ex 2 1
0
at x = 0, we are faced with f (0) 5
x
0
which is undefined. However, if we start with, say, x = ±1, and allow x to approach 0
If we attempt to evaluate the function f(x) 5
from either direction, we can see that the value is in fact 1. This is illustrated in the table
of values below, which shows that as x gets closer and closer to 0 (through both positive
ex 2 1
51
0
x
and negative values), the function approaches 1. In symbols, lim
x
x
1
0.1
0.01
0.001
f (x)
1.71828
1.05171
1.00501
1.00050
x
−1
f(x)
0.63212
−0.1
0.95163
−0.01
0.99502
−0.001
0.99950
A graph of f(x) is sketched in Figure 2.18. Although the function is undefined at x = 0, the
graph shows that it is possible to ‘plug the hole’ by defining f(0) = 1.
y
ex 2 1
f(x) 5
x
1
x
Figure 2.18
188 Chapter 2 Non-linear Equations
Definition A function f is continuous at x = a if lim f (x) 5 f (a)
x
a
In other words, a function is continuous at x = a provided the function can be evaluated
at x = a (so that f (a) exists), and that the limit also exists and equals f (a).
Referring to the functions illustrated in Figure 2.17 and 2.18, we see that:
●
The function in Figure 2.17(b) is discontinuous because f is undefined at x = 2.
●
The function in Figure 2.17(c) is discontinuous because the lim f (x) does not exist (it gives
x
2
a value 5 as x approaches 2 from the right but a different value of 3 as x approaches 2 from
the left).
●
The function in Figure 2.18 is discontinuous because f is undefined at x = 0.
It is possible to remove the discontinuity for the function, f(x) 5
ex 2 1
5 1. If we are prepared to define the function piecewise as
0
x
previous example, lim
x
ex 2 1
. As we saw in the
x
ex 2 1
if x Þ 0
f (x) 5
x
1 if x 5 0
then the function is now defined at x = 0 and the value of the limit and the function both give
the value of 1 at this point.
Continuous functions have many important properties. In particular, it is necessary for a
function to be continuous in order even to consider the idea of differentiability which we will
investigate in Chapter 4.
Key Terms
Continuous The name given to a function which can be drawn without taking a pen off the paper.
More formally, when lim f (x) 5 f (a) at all points in the domain.
x
Degree
a
The highest power in a polynomial.
Discontinuous The name given to a function which is not continuous everywhere. The graph of
the function has jumps or gaps.
Polynomial A function of the form anx n + an−1x n−1 + . . . + a0.
Multiple choice questions
Question 1
Simplify
2x23y5/2
6x25y
A.
x28y3/2
3
B
x2y3/2
3
C.
x2y5/2
3
D.
x28y7/2
3
E.
x28y5/2
3
Question 2
Use the quadratic formula to solve
3x2 = 7x − 1
A.
27 6 61
6
B
27 6 37
6
C
7 6 61
6
D.
7 6 37
6
E No real solutions
Question 3
If fixed costs are 20, variable costs per unit are 2 and the demand function is P + 2Q = 24,
obtain an expression for the profit function in terms of Q.
A. −2Q2 + 20Q − 2
B. −2Q2 + 9Q − 10
C. −2Q2 + 22Q − 20
D. −2Q2 + 26Q + 20
E. −2Q2 + 22Q − 10
190 Chapter 2 Non-linear Equations
Question 4
The graph shows the total revenue function and total cost function plotted on the same
diagram. Use the graph to estimate the maximum profit.
6000
5000
4000
3000
2000
1000
20
40
60
80
100
Quantity
A. 4700
B. 2100
C. 5000
D. 1800
E. 2500
Question 5
If fixed costs are $80 000 and variable costs are $300 per item, how many items can be
produced for $123 800?
A. 152
B. 150
C. 146
D. 172
E. 164
Question 6
Which one of the following statements best describes the homogeneity of the production
function f(K, L) 5 8 L2K 1 2KL?
A. It displays decreasing returns to scale
B. It displays increasing returns to scale
C. It displays constant returns to scale
D. It is not homogeneous
E. There is insufficient information
Multiple choice questions 191
Question 7
A firm’s fixed costs are 50 and the variable costs per unit are 3Q. Draw a sketch graph of
the average cost function.
AC
AC
Q
A.
Q
B.
AC
AC
Q
C.
Q
D.
AC
Q
E.
Question 8
Simplify loga 14 − loga 7 + 2loga x − 3loga y − loga 1
A. loga
2x2
y3
B. loga
7x2
3y
C. loga
y3
2x2
D. loga
7x2
y3
E. loga
y2
7x3
192 Chapter 2 Non-linear Equations
Question 9
If one solution of the quadratic equation, 21x2 + cx − 10 = 0 is 2/7, find the second ­solution.
A. −5/3
B. −7/2
C. −2/7
D. −3/7
E. −3/5
Question 10
A firm’s demand function is 3P + 2Q = 60. Fixed costs are 13 and the variable cost per
unit is 4. At what quantities does the firm make a profit of 77?
A. 5 and 27
B. 9 and 15
C. 8 and 20
D. 6 and 18
E. 4 and 16
Question 11
The demand function of a good is P = 15 − 2Q − Q2. Find the largest possible value of Q.
A. 5
C. 7.5
B. 15
D. 3
E. 8
Question 12
The value V (in billions of $) of a non-renewable resource t years from now is given by
V(t) = Ae0.05t. If it is currently worth $50 billion, after how many years will it be worth $175
billion?
A. 20 ln 2
7
2
B. 2
1
2
ln
20
7
C. 20 ln
2
7
D. 20.05ln
7
2
E. 20 ln
7
2
Question 13
Given the supply and demand functions
P = Q2S + 8QS + 25
P = −Q2D − 7QD + 273
determine the equilibrium price.
A. 15.5
B. 8
C. 105
D. 153
E. 97
Question 14
Solve the quadratic inequality −Q2 + 7Q ≥ 10.
A. Q ≤ 2, Q ≥ 5
B 2≤Q≤5
D. −5 ≤ Q ≤ −2
E No solution
C Q ≤ −5, Q ≥ −2
Question 15
Find the exact solution of the equation 18e−3x = 2e−5x
A. −ln 3
B. 2 ln 3
C.
5
ln 2
3
D.
3
ln 2
5
3
E. 2 ln 2
5
Examination questions
Question 1
The output Q of a production process can be modelled by the Cobb-Douglas function,
Q = 100K2/3 L1/2, where K and L denote the inputs of capital and labour, respectively.
(a) Find Q when K = 8 and L = 25.
(b) Find L when Q = 9000 and K = 27.
(c) Find K when Q = 3200 and L = 4.
(d) Show that this function is homogeneous and comment on its return to scale.
Question 2
A firm’s fixed costs are 8, variable costs per unit are 2 and the demand function is
P = 8 − Q where P and Q denote price and quantity, respectively.
Find an expression for profit, π, in terms of Q and hence sketch a graph of π against Q.
(a) For what values of Q does the firm break-even?
(b) Calculate the maximum profit.
(c) Describe the effect on the graph of the profit function due to a decrease in fixed
costs.
Question 3
The supply and demand functions of a good are given by:
P = 2Q2S + 3QS + 10
P = −Q2D − 3QD + 154
where P, QS and QD denote price, quantity supplied and quantity demanded, respectively.
(a) Determine the equilibrium price and quantity.
(b) If the government imposes a tax of $5 per item, determine the new values of equilibrium
price and quantity. Round your answers to two decimal places. Who pays the tax?
Question 4
The demand function of a good is P = 10 − Q and the total cost function is TC = 12 + 2Q.
(a) Find an expression for the total revenue function and deduce that the profit function is
given by π = −Q2 + 8Q − 12.
(b) By solving the quadratic inequality π ≥ 3, determine the range of prices which will
ensure that the firm’s profit is at least 3 units.
194 Chapter 2 Non-linear Equations
Question 5
(a) Giving FULL details of your working, use the rules of logs to:
(i) express 5loga x − 2loga y + loga xy as a single logarithm
1
(ii) evaluate log2 5 1 log2 36 2 log2 120
2
(b) The price of a good over a three-year period is shown in the following table:
Beginning of Year
Price (in $)
0
1
2
1000
1100
1210
Explain why this data can be modelled exactly using an exponential function, p = Abt
where p denotes the price at the beginning of year t.
State the values of the parameters, A and b.
Assuming that the model continues to be valid in subsequent years, work out the
(i) price of the good (to the nearest $) at the beginning of year 10;
(ii) number of complete years for the price of the good to exceed $8000.
Question 6
A firm’s fixed costs are 14 and the variable costs are 2 per unit. The corresponding profit
function is of the form π = aQ2 + bQ + c where Q denotes quantity.
(a) Write down expressions for the total cost and average cost functions, and sketch their
graphs.
(b) Explain why c = −14.
(c) The firm makes a loss of 6 when Q = 1 and a profit of 4 when Q = 6. Write down a
pair of simultaneous equations and solve them to find a and b.
(d) Find the break-even points and hence find the maximum profit.
(e) Find an expression for total revenue and deduce the demand function.
Question 7
(a) Simplify
(i)
3a2b4
6a23b
(ii) (8x3 y6)2/3
(iii) logb b6
(iv) logb xy 1 logb xz 2 logb
x
yz
(b) Solve the equation 3x = 2(5x), giving your answer correct to two decimal places.
(c) The total number of sales, S, made t weeks after the launch of a new model of phone
is given by
S = 2 000 000(1 − e−0.2t)
(i) Estimate the number of sales after two weeks.
(ii) How many weeks does it take for sales to exceed one million? Give your answer
correct to one decimal place.
(iii) Sketch a graph of S against t to show how sales vary over time.
Examination questions 195
Question 8
The value, V, of a piece of machinery depreciates at a rate which is proportional to V.
Two years after purchase, the machinery is worth $45 125, and four years after purchase,
it is worth $40 725.
(a) If V = Abt where t is the number of years after purchase, find A and b, and explain the
significance of the number A.
(b) By using your answer to part (a) or otherwise, express V in the form Aert where r is
given to three significant figures.
(c) If a graph of ln V is plotted against t, show that the graph will be a straight line, and
state the gradient and vertical intercept.
Question 9
A hospital administrator decides to review the costs of three departments of the organisation which are summarised in the table. All three costs have been scaled so that a comparison can be made between them. The variable Q is measured in thousands of patients.
Fixed
Variable
IT
Retail
Printing
1
5
0
0.04Q + 0.5
0.5
1
(a) Derive expressions for the average cost functions for each of these departments.
(b) The diagram shows the graphs of all three average cost functions on the same diagram.
Which hospital department corresponds to each graph? Justify your answer.
AC
A
B
C
Q
(c) Find the values of Q for which the average cost of IT is less than the average cost of
printing.
Chapter 3
Mathematics of Finance
This chapter provides an understanding of the way in which financial calculations are worked
out. There are four sections, which should be read in the order that they appear.
Section 3.1 revises work on percentages. In particular, a quick method of dealing with percentage increase and decrease calculations is described. This enables an overall percentage change
to be deduced easily from a sequence of individual changes. Percentages are used to calculate
and interpret index numbers, and to adjust value data for inflation.
Section 3.2 shows how to calculate the future value of a lump sum which is invested to earn
interest. This interest can be added to the investment annually, semi-annually, quarterly or even
more frequently. The exponential function is used to solve problems in which interest is compounded continuously.
A wide variety of applications are considered in Sections 3.3 and 3.4. In Section 3.3 a mathematical device known as a geometric progression, which is used to calculate the future value of a
savings plan and the monthly repayments of a loan, is introduced. Section 3.4 describes the
opposite problem of calculating the present value given a future value. The process of working
backwards is called discounting. It can be used to decide how much money to invest today in
order to achieve a specific target sum in a few years’ time. Discounting can be used to appraise
different investment projects. On the macroeconomic level, the relationship between interest
rates and speculative demand for money is investigated.
The material in this chapter will be of greatest benefit to students on business studies and
management courses. This chapter could be omitted without affecting your understanding of
the rest of this text.
Section 3.1
Percentages
Objectives
At the end of this section you should be able to:
●
Understand what a percentage is.
●
Solve problems involving a percentage increase or decrease.
●
Write down scale factors associated with percentage changes.
●
Work out overall percentage changes.
●
Calculate and interpret index numbers.
●
Adjust value data for inflation.
Advice
The first part of this section provides a leisurely revision of the idea of a percentage as well as
a reminder about how to use scale factors to cope with percentage changes. These ideas are crucial
to any understanding of financial mathematics. However, if you are already confident in using
percentages, you may wish to move straight on to the applications covered in sub-sections 3.1.1
and 3.1.2.
In order to be able to handle financial calculations, it is necessary to use percentages profi­
ciently. The word ‘percentage’ literally means ‘per cent’, i.e. per hundredth, so that whenever
we speak of r% of something, we simply mean the fraction (r/100) ths of it.
For example,
25% is the same as
25
1
5
100 4
30% is the same as
30
3
5
100 10
50% is the same as
50
1
5
100 2
Example
Calculate
(a) 15% of 12
(b) 150% of 290
Section 3.1 Percentages 199
Solution
(a) 15% of 12 is the same as
15
3 12 5 0.15 3 12 5 1.8
100
(b) 150% of 290 is the same as
150
3 290 5 1.5 3 290 5 435
100
Practice Problem
1. Calculate
(a) 10% of $2.90 (b) 75% of $1250 (c) 24% of $580
Whenever any numerical quantity increases or decreases, it is customary to refer to this
change in percentage terms. The following example serves to remind you of how to perform
calculations involving percentage changes.
Example
(a) An investment rises from $2500 to $3375. Express the increase as a percentage of the
original.
(b) At the beginning of a year, the population of a small village is 8400. If the annual rise
in population is 12%, find the population at the end of the year.
(c) In a sale, all prices are reduced by 20%. Find the sale price of a good originally costing
$580.
Solution
(a) The rise in the value of the investment is
3375 − 2500 = 875
As a fraction of the original, this is
875
5 0.35
2500
This is the same as 35 hundredths, so the percentage rise is 35%.
➜
200 Chapter 3 Mathematics of Finance
(b) As a fraction,
12% is the same as
12
5 0.12
100
so the rise in population is
0.12 × 8400 = 1008
Hence the final population is
8400 + 1008 = 9408
(c) As a fraction,
20% is the same as
20
5 0.2
100
so the fall in price is
0.2 × 580 = 116
Hence the final price is
580 − 116 = $464
Practice Problem
2. (a) A firm’s annual sales rise from 50 000 to 55 000 from one year to the next. Express
the rise as a percentage of the original.
(b) The government imposes a 15% tax on the price of a good. How much does the
consumer pay for a good priced by a firm at $1360?
(c) Investments fall during the course of a year by 7%. Find the value of an investment
at the end of the year if it was worth $9500 at the beginning of the year.
In the previous example and in Practice Problem 2, the calculations were performed in two
separate stages. The actual rise or fall was first worked out, and these changes were then
applied to the original value to obtain the final answer. It is possible to obtain this answer
in a single calculation, and we now describe how this can be done. Not only is this new
approach quicker, but it also enables us to tackle more difficult problems. To be specific,
let us suppose that the price of a good is set to rise by 9% and that its current price is $78.
The new price consists of the original (which can be thought of as 100% of the $78) plus
the increase (which is 9% of $78). The final price is therefore
100% + 9% = 109% (of the $78)
which is the same as
109
5 1.09
100
Section 3.1 Percentages 201
In other words, in order to calculate the final price, all we have to do is to multiply by the
scale factor, 1.09. Hence the new price is
1.09 × 78 = $85.02
One advantage of this approach is that it is then just as easy to go backwards and work
out the original price from the new price. To go backwards in time we simply divide by the
scale factor. For example, if the final price of a good is $1068.20, then before a 9% increase
the price would have been
1068.20 ÷ 1.09 = $980
In general, if the percentage rise is r%, then the final value consists of the original (100%)
together with the increase (r%), giving a total of
100
r
r
1
511
100 100
100
To go forwards in time we multiply by this scale factor, whereas to go backwards we divide.
Example
(a) If the annual rate of inflation is 4%, find the price of a good at the end of a year if its
price at the beginning of the year is $25.
(b) The cost of a good is $750 including a sales tax of 20%. What is the cost excluding the
sales tax?
(c) Express the rise from 950 to 1007 as a percentage.
Solution
(a) The scale factor is
11
4
5 1.04
100
We are trying to find the price after the increase, so we multiply to get
25 × 1.04 = $26
(b) The scale factor is
11
20
5 1.2
100
This time we are trying to find the price before the increase, so we divide by the scale
factor to get
750 ÷ 1.2 = $625
(c) The scale factor is
new value 1007
5
5 1.06
950
old value
which can be thought of as
11
6
100
so the rise is 6%.
202 Chapter 3 Mathematics of Finance
Practice Problem
3. (a) The value of a good rises by 13% in a year. If it was worth $6.5 million at the
beginning of the year, find its value at the end of the year.
(b) The GNP of a country has increased by 63% over the past five years and is now
$124 billion. What was the GNP five years ago?
(c) Sales rise from 115 000 to 123 050 in a year. Find the annual percentage rise.
It is possible to use scale factors to solve problems involving percentage decreases. To be
specific, suppose that an investment of $76 falls by 20%. The new value is the original (100%)
less the decrease (20%), so it is 80% of the original. The scale factor is therefore 0.8, giving
a new value of
0.8 × 76 = $60.80
In general, the scale factor for an r% decrease is
100
r
r
2
512
100 100
100
Once again, you multiply by this scale factor when going forwards in time and divide when
going backwards.
Example
(a) The value of a car depreciates by 25% in a year. What will a car, currently priced at
$43 000, be worth in a year’s time?
(b) After a 15% reduction in a sale, the price of a good is $39.95. What was the price before
the sale began?
(c) The number of passengers using a rail link fell from 190 205 to 174 989. Find the
percentage decrease.
Solution
(a) The scale factor is
12
25
5 0.75
100
so the new price is
43 000 × 0.75 = $32 250
forwards in time
so multiply
(b) The scale factor is
12
15
5 0.85
100
so the original price was
39.95 ÷ 0.85 = $47
backwards in time
so divide
Section 3.1 Percentages 203
(c) The scale factor is
new value 174 989
5
5 0.92
190 205
old value
which can be thought of as
12
8
100
so the fall is 8%.
not 92%!
Practice Problem
4. (a) Current monthly output from a factory is 25 000. In a recession, this is expected to
fall by 65%. Estimate the new level of output.
(b) As a result of a modernisation programme, a firm is able to reduce the size of its
workforce by 24%. If it now employs 570 workers, how many people did it employ
before restructuring?
(c) Shares originally worth $10.50 fall in a stock market crash to $2.10. Find the
percentage decrease.
The final application of scale factors that we consider is to the calculation of overall
percentage changes. It is often the case that over various periods of time, the price of a good
is subject to several individual percentage changes. It is useful to be able to replace these
by an equivalent single percentage change spanning the entire period. This can be done by
simply multiplying together successive scale factors.
Example
(a) Share prices rise by 32% during the first half of the year and rise by a further 10%
during the second half. What is the overall percentage change?
(b) Find the overall percentage change in the price of a good if it rises by 5% in a year but
is then reduced by 30% in a sale.
Solution
(a) To find the value of shares at the end of the first 6 months, we would multiply by
11
32
5 1.32
100
and at the end of the year we would multiply again by the scale factor
11
10
5 1.1
100
➜
204 Chapter 3 Mathematics of Finance
The net effect is to multiply by their product
1.32 × 1.1 = 1.452
which can be thought of as
11
45.2
100
so the overall change is 45.2%.
Notice that this is not the same as
32% + 10% = 42%
This is because during the second half of the year, we not only get a 10% rise in the
original value, but we also get a 10% rise on the gain accrued during the first 6 months.
(b) The individual scale factors are 1.05 and 0.7, so the overall scale factor is
1.05 × 0.7 = 0.735
The fact that this is less than 1 indicates that the overall change is a decrease. Writing
0.735 5 1 2 0.265 5 1 2
26.5
100
we see that this scale factor represents a 26.5% decrease.
Practice Problem
5. Find the single percentage increase or decrease equivalent to
(a) an increase of 30% followed by an increase of 40%
(b) a decrease of 30% followed by a decrease of 40%
(c) an increase of 10% followed by a decrease of 50%.
We conclude this section by describing two applications of percentages in macroeconomics:
●
index numbers;
●
inflation.
We consider each of these in turn.
3.1.1 Index numbers
Economic data often take the form of a time series; values of economic indicators are avail­
able on an annual, quarterly or monthly basis, and we are interested in analysing the rise and
fall of these numbers over time. Index numbers enable us to identify trends and relationships
in the data. The following example shows you how to calculate index numbers and how to
interpret them.
Section 3.1 Percentages 205
Example
Table 3.1 shows the values of household spending (in billions of dollars) during a five-year
period. Calculate the index numbers when Year 2 is taken as the base year and give a brief
interpretation.
Table 3.1
Year
Household spending
1
2
3
4
5
686.9
697.2
723.7
716.6
734.5
Solution
When finding index numbers, a base year is chosen and the value of 100 is allocated to that
year. In this example, we are told to take Y2 as the base year, so the index number
of Y2 is 100. To find the index number of the year Y3, we work out the scale factor asso­
ciated with the change in household spending from the base year, Y2 to Y3, and then
multiply the answer by 100.
index number 5 scale factor from base year 3 100
In this case, we get
723.7
3 100 5 103.8
697.2
This shows that the value of household spending in Y3 was 103.8% of its value in Y2. In
other words, household spending increased by 3.8% during Y3.
For the year Y4, the value of household spending was 716.6, giving the index number
716.6
3 100 5 102.8
697.2
This shows that the value of household spending in Y4 was 102.8% of its value in Y2.
In other words, household spending increased by 2.8% between Y2 and Y4. Notice that
this is less than that calculated for Y3, reflecting the fact that spending actually fell slightly
during Y4. The remaining two index numbers are calculated in a similar way and are
shown in Table 3.2.
Table 3.2
Year
Household spending
Index number
1
2
3
4
5
686.9
98.5
697.2
100
723.7
103.8
716.6
102.8
734.5
105.3
Practice Problem
6. Calculate the index numbers for the data shown in Table 3.1, this time taking Year 1
as the base year.
206 Chapter 3 Mathematics of Finance
Index numbers themselves have no units. They merely express the value of some quantity
as a percentage of a base number. This is particularly useful, since it enables us to compare
how values of quantities, of varying magnitudes, change in relation to each other. Table 3.3
shows the rise and fall of two share prices during an 8-month period. The prices (in dollars)
listed for each share are taken on the last day of each month. Share A is exceptionally cheap.
Investors often include this type of share in their portfolio, since they can occasionally make
spectacular gains. This was the case with many dot.com shares at the end of the 1990s. The
second share is more expensive and corresponds to a larger, more established firm.
Table 3.3
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Share A
Share B
0.31
6.34
0.28
6.40
0.31
6.45
0.34
6.52
0.40
6.57
0.39
6.43
0.45
6.65
0.52
7.00
The index numbers have been listed in Table 3.4, taking April as the base month. Notice
that both shares are given the same index number of 100 in April. This is despite the fact that
the values of the two shares are very different. This creates ‘a level playing-field’, enabling us
to monitor the relative performance of the two shares. The index numbers show quite clearly
that share A has outperformed share B during this period. Indeed, if an investor had spent
$1000 on shares of type A in January, they could have bought 3225 of them, which would be
worth $1677 in August, making a profit of $677. The corresponding profit for share B is only $103.
Table 3.4
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Index of share price A (April = 100)
Index of share price B (April = 100)
91.2
97.2
82.3
98.2
91.2
98.9
100
100
117.6
100.8
114.7
98.6
132.4
102.0
152.9
107.4
Incidentally, if the only information you have about the time series is the set of index
numbers, then it is possible to work out the percentage changes between any pair of values.
Table 3.5 shows the index numbers of the output of a particular firm for two consecutive
years.
Table 3.5
Output
Index
Y1Q1
Y1Q2
Y1Q3
Y1Q4
Y2Q1
Y2Q2
Y2Q3
Y2Q4
89.3
98.1
105.0
99.3
100
106.3
110.2
105.7
The table shows that the base quarter is the first quarter of Year 2 because the index num­
ber is 100 in Y2Q1. It is, of course, easy to find the percentage change from this quarter to any
subsequent quarter. For example, the index number associated with the third quarter of Y2
is 110.2, so we know immediately that the percentage change in output from Y2Q1 to Y2Q3
is 10.2%. However, it is not immediately obvious what the percentage change is from, say,
Y1Q2 to Y2Q2. To work this out, note that the scale factor of this change is
106.3
5 1.084
98.1
which corresponds to an 8.4% increase.
Section 3.1 Percentages 207
Similarly, the scale factor of the change from Y1Q3 to Y2Q1 is
100
5 0.952
105
This is less than 1, reflecting the fact that output has fallen. To find the percentage change, we
write the scale factor as
1 − 0.048
which shows that the percentage decrease is 4.8%.
Practice Problem
7. Use the index numbers listed in Table 3.5 to find the percentage change in output from
(a) Y2Q1 to Y2Q4
(b) Y1Q1 to Y2Q4
(c) Y1Q1 to Y2Q1
It is possible to create sensible index numbers to measure the variation of a bundle of goods
over time. To be specific, suppose that a firm buys three products. Table 3.6 shows the number
of each type bought in Year 1 together with the unit prices of each item in Year 1 and Year 2.
Table 3.6
Product
Number bought in Year 1
Unit price in Year 1
Unit price in Year 2
A
B
C
20
35
10
8
18
6
10
23
5
The total purchase cost in Year 1 is worked out by multiplying the quantities by the corres­
ponding prices which gives
20 × 8 + 35 × 18 + 10 × 6 = 850
If we assume that the firm buys the same number of each type in Year 2, the total cost would be
20 × 10 + 35 × 23 + 10 × 5 = 1055
Taking the base year as Year 1, we see that the index number in Year 1 is 100 and the index
number in Year 2 is
1055
3 100 5 124.1
850
In this calculation we have created a weighted index which assumes that the quantities
purchased in Year 2 are the same as the base year. This is referred to as a base-weighted index
(or Laspeyres index). It provides a good indication to the firm of changing costs, although
it becomes unreliable if the firm changes the amount it buys over time. If figures are available
for the quantities purchased each year, then it is possible to work out an index based on
current values. For example, if the quantities bought in Year 2 are those shown in Table 3.7,
then the current weighted index in Year 2 is
17 3 10 1 38 3 23 1 12 3 5
1104
3 100 5
3 100 5 129.9
850
850
208 Chapter 3 Mathematics of Finance
Table 3.7
Product
Number bought in Year 2
A
B
C
17
38
12
This is referred to as the Paasche index. It has the obvious advantage of reflecting changes in
the actual purchases made by the firm. On the downside, it is necessary to know the precise
amounts of each item bought every year and, strictly speaking, since the calculations change
each year, you are not comparing like quantities.
3.1.2 Inflation
Over a period of time, the prices of many goods and services usually increase. The annual
rate of inflation is the average percentage change in a given selection of these goods and ser­
vices, over the previous year. Seasonal variations are taken into account, and the particular
basket of goods and services is changed periodically to reflect changing patterns of household
expenditure. The presence of inflation is particularly irritating when trying to interpret a time
series that involves a monetary value. Inevitably, this will be influenced by inflation during
any year, and what is of interest is the fluctuation of a time series ‘over and above’ inflation.
Economists deal with this by distinguishing between nominal and real data. Nominal data are
the original, raw data such as those listed in tables in the previous sub-section. These are
based on the prices that prevailed at the time. Real data are the values that have been adjusted
to take inflation into account. The standard way of doing this is to pick a year and then con­
vert the values for all other years to the level that they would have had in this base year. This
may sound rather complicated, but the idea and calculations involved are really quite simple
as the following example demonstrates.
Example
Table 3.8 shows the price (in thousands of dollars) of an average house in a certain town
during a five-year period. The price quoted is the value of the house at the end of each
year. Use the annual rates of inflation given in Table 3.9 to adjust the prices to those pre­
vailing at the end of Year 2. Compare the rise in both the nominal and real values of house
prices during this period.
Table 3.8
Year
Average house price
1
2
3
4
5
72
89
93
100
106
Table 3.9
Year
Annual rate of inflation
2
3
4
5
10.7%
7.1%
3.5%
2.3%
Section 3.1 Percentages 209
Solution
The raw figures shown in Table 3.8 give the impression that houses increased steadily in
value throughout this period, with a quite substantial gain during the first year. However,
if inflation had been very high, then the gain in real terms would have been quite small.
Indeed, if the rate of inflation were to exceed the percentage rise of these nominal data,
then the price of a house would actually fall in real terms. To analyse this situation we will
use Table 3.9, which shows the rates of inflation during this period. Notice that since the
house prices listed in Table 3.9 are quoted at the end of each year, we are not interested in
the rate of inflation during Year 1.
We are told in the question to choose Year 2 as the base year and calculate the value of
the house at ‘Year 2 prices’. The value of the house at the end of Year 2 is obviously $89 000,
since no adjustment needs to be made. At the end of Year 3, the house is worth $93 000.
However, during that year, inflation was 7.1%. To adjust this price to ‘Year 2 prices’ we
simply divide by the scale factor 1.071, since we are going backwards in time. We get
93 000
5 86 835
1.071
In real terms the house has fallen in value by over $2000.
To adjust the price of the house in Year 4 we first need to divide by 1.035 to backtrack
to Year 3, and then divide again by 1.071 to reach Year 2. We get
100 000
5 90 213
1.035 3 1.071
In real terms there has at least been some gain during Year 4. However, this is less than
impress­ive, and from a purely financial point of view, there would have been more
lucrative ways of investing this capital.
For the Year 5 price, the adjusted value is
106 000
5 93 476
1.023 3 1.035 3 1.071
and, for Year 1, the adjusted value is
72 000 × 1.107 = 79 704
going forward in
time, so multiply
Table 3.10 lists both the nominal and the ‘constant Year 2’ values of the house (rounded
to the nearest thousand) for comparison. It shows quite clearly that, apart from the gain
during Year 2, the increase in value has, in fact, been quite modest.
Table 3.10
Year
Nominal house price
Year 2 house price
1
2
3
4
5
72
80
89
89
94
87
100
90
106
93
210 Chapter 3 Mathematics of Finance
Practice Problem
8. Table 3.11 shows the average annual salary (in thousands of dollars) of employees in
a small firm, together with the annual rate of inflation for that year. Adjust these salaries
to the prices prevailing at the end of Year 2 and so give the real values of the employees’
salaries at constant ‘Year 2 prices’. Comment on the rise in earnings during this period.
Table 3.11
Year
Salary
Inflation
1
2
3
4
5
17.3
18.1
4.9
19.8
4.3
23.5
4.0
26.0
3.5
Key Terms
Index number The scale factor of a variable measured from the base year multiplied by 100.
Inflation
The percentage increase in the level of prices over a 12-month period.
Laspeyre index An index number for groups of data which are weighted by the quantities used in
the base year.
Nominal data Monetary values prevailing at the time that they were measured.
Paasche index An index number for groups of data which are weighted by the quantities used in
the current year.
Real data
Monetary values adjusted to take inflation into account.
Scale factor The multiplier that gives the final value in percentage problems.
Time series
A sequence of numbers indicating the variation of data over time.
Exercise 3.1
1. Express the following percentages as fractions in their simplest form:
(a) 35%
(b) 88%
(c) 250%
(d) 171/2%
(e) 0.2%
2. Calculate each of the following:
(a) 5% of 24 (b) 8% of 88 (c) 48% of 4563 (d) 112% of 56
3. A firm has 132 female and 88 male employees.
(a) What percentage of staff are female?
(b) During the next year 8 additional female staff are employed. If the percentage of
female staff is now 56%, how many additional male staff were recruited during the
year?
Section 3.1 Percentages 211
4. Write down the scale factors corresponding to
(a) an increase of 19%;
(b) an increase of 250%;
(c) a decrease of 2%;
(d) a decrease of 43%.
5. Write down the percentage changes corresponding to the following scale factors:
(a) 1.04
(b) 1.42
(c) 0.86
(d) 3.45
(e) 1.0025
(f) 0.04
6. Find the new quantities when
(a) $16.25 is increased by 12%
(b) the population of a town, currently at 113 566, rises by 5%
(c) a good priced by a firm at $87.90 is subject to a sales tax of 15%
(d) a good priced at $2300 is reduced by 30% in a sale
(e) a car, valued at $23 000, depreciates by 32%
7. A student discount card reduces a bill in a restaurant from $124 to $80.60. Work out
the percentage discount.
8. A shop sells books at ‘20% below the recommended retail price (r.r.p.)’. If it sells a
book for $12.40, find
(a) the r.r.p.
(b) the cost of the book after a further reduction of 15% in a sale
(c) the overall percentage discount obtained by buying the book from the shop in the
sale compared with the manufacturer’s r.r.p.
9. A TV costs $900 including 20% sales tax. Find the new price if tax is reduced to 15%.
10. An antiques dealer tries to sell a vase at 45% above the $18 000 which the dealer paid
at auction.
(a) What is the new sale price?
(b) By what percentage can the dealer now reduce the price before making a loss?
11. Find the single percentage increase or decrease equivalent to
(a) a 10% increase followed by a 25% increase
(b) a 34% decrease followed by a 65% increase
(c) a 25% increase followed by a 25% decrease.
Explain in words why the overall change in part (c) is not 0%.
12. Table 3.12 gives the annual rate of inflation during a five-year period.
Table 3.12
Year
1
2
3
4
5
Annual rate of inflation
1.8%
2.1%
2.9%
2.4%
2.7%
➜
212 Chapter 3 Mathematics of Finance
If a nominal house price at the end of Year 1 was $10.8 million, find the real house price
adjusted to prices prevailing at the end of Year 4. Round your answer to three signifi­
cant figures.
13. The price of a good during the last five years is:
$25 $30 $36 $43 $50
Calculate the index numbers using the last year as the base year and hence comment on
the rise in prices during this period.
14. Table 3.13 shows the monthly index of sales of a good during the first four months of
the year.
Table 3.13
Month
Jan
Feb
Mar
Apr
Index
100
120
145
150
(a) Which month is chosen as the base year?
(b) If sales in February are 3840, what are the sales in April?
(c) What is the index number in May if sales are 4256?
15. Table 3.14 shows the index numbers associated with transport costs during a 20-year
period. The public transport costs reflect changes to bus and train fares, whereas pri­
vate transport costs include purchase, service, petrol, tax and insurance costs of cars.
Table 3.14
Year
Public transport
Private transport
1995
2000
2005
2010
2015
100
100
130
125
198
180
224
199
245
221
(1) Which year is chosen as the base year?
(2) Find the percentage increases in the cost of public transport from
(a) 1995 to 2000 (b) 2000 to 2005 (c) 2005 to 2010 (d) 2010 to 2015
(3) Repeat part (2) for private transport.
(4) Comment briefly on the relative rise in public and private transport costs during this
20-year period.
16. Table 3.15 shows the number of items (in thousands) produced from a factory pro­
duction line during the course of a year. Taking the second quarter as the base quarter,
calculate the associated index numbers. Suggest a possible reason for the fluctuations
in output.
Table 3.15
Quarter
Output
Q1
Q2
Q3
Q4
13.5
1.4
2.5
10.5
Section 3.1 Percentages 213
17. Table 3.16 shows the prices of a good for each year between Year 1 and Year 6.
Table 3.16
Year
1
2
3
4
5
6
Price ($)
40
48
44
56
60
71
(a) Work out the index numbers, correct to one decimal place, taking Year 2 as the base
year.
(b) If the index number for Year 7 is 135, calculate the corresponding price. You may
assume that the base year is still Year 2.
(c) If the index number in Year 3 is approximately 73, find the year that is used as the
base year.
Exercise 3.1*
1. Total revenue from daily ticket sales to a theme park is $1 352 400. A total of 12 000
tickets were sold and 65% of these were child’s tickets with a 30% discount off the
adult price. Work out the cost of an adult ticket.
2. The cost of a computer is $6000 including 20% sales tax. In a generous gesture, the
govern­ment decides to reduce the rate to just 17.5%. Find the cost of the computer after
the tax has changed.
3. A coat originally costing $150 is reduced by 25% in a sale and, since nobody bought
the coat, a further reduction of 20% of the sale price is applied.
(a) Find the final cost of the coat after both reductions.
(b) Find the overall percentage reduction, and explain why this is not the same as a
single reduction of 45%.
4. A furniture store has a sale of 40% off on selected items.
A sales assistant, Carol, reduces the price of a sofa originally costing $1200.
(a) What is the new price?
The manager does not want this sofa to be in the sale and the following day tells
another sales assistant, Michael, to restore the sofa back to the original price. He
does not know what the original price was and decides to show off his mathematical
knowledge by taking the answer to part (a) and multiplying it by 1.4.
(b) Explain carefully why this does not give the correct answer of $1200.
(c) Suggest an alternative calculation that would give the right answer.
5. During 2016 the price of a good increased by 8%. In the sales on 1 January 2017 all
items are reduced by 25%.
(a) If the sale price of the good is $688.50, find the original price at the beginning of
2014.
(b) Find the overall percentage change.
(c) What percentage increase would be needed to restore the cost to the original price
prevailing on 1 January 2016? Give your answer to one decimal place.
➜
214 Chapter 3 Mathematics of Finance
6. Table 3.17 shows government expenditure (in billions of dollars) on education for four
consecutive years, together with the rate of inflation for each year.
(a) Taking Year 1 as the base year, work out the index numbers of the nominal data
given in the third row of the table.
(b) Find the values of expenditure at constant Year 1 prices and hence recalculate the
index numbers of real government expenditure.
(c) Give an interpretation of the index numbers calculated in part (b).
Table 3.17
Year
Spending
Inflation
1
2
3
4
236
240
4.7
267
4.2
276
3.4
7. Index numbers associated with the growth of unemployment during an eight-year
period are shown in Table 3.18.
(a) What are the base years for the two indices?
(b) If the government had not switched to index 2, what would be the values of index
1 in years 7 and 8?
(c) What values would index 2 have been in years 1, 2, 3, 4 and 5?
(d) If unemployment was 1.2 million in year 4, how many people were unemployed in
years 1 and 8?
Table 3.18
Year
Index 1
Index 2
1
2
3
4
5
6
7
8
100
95
105
110
119
127
100
112
118
8. The prices of a good at the end of each year between Year 1 and Year 6 are listed in
Table 3.19, which also shows the annual rate of inflation.
Table 3.19
Year
1
2
3
4
5
6
Price
Inflation
230
242
4%
251
3%
257
2.5%
270
2%
284
2%
(a) Find the values of the prices adjusted to the end of Year 2, correct to two decimal
places. Hence, calculate the index numbers of the real data with Year 2 as the base
year. Give your answers correct to one decimal place.
(b) If the index number of the real price for Year 7 is 109 and the rate of inflation for
that year is 2.5%, work out the nominal value of the price in Year 7. Give your
answer rounded to the nearest whole number.
(c) If the index number of the real data in Year 0 is 95.6 and the nominal price is $215,
find the rate of inflation for Year 1. Give your answer correct to one decimal place.
Section 3.1 Percentages 215
9. A firm buys three goods. The number of each type bought in Year 1 is shown in Table 3.20
together with the unit prices for three consecutive years.
Table 3.20
Product
Number bought in Year 1
Unit price in Year 1
Unit price in Year 2
Unit Price in Year 3
A
B
C
56
40
122
34
24
13
36
24
11
42
23
14
(a) Calculate a base-weighted index (Laspeyres) for these data using Year 1 as the base
year.
(b) Comment on the values obtained in part (a).
10. The number of goods bought by the firm in Q9 vary over time. Figures for Year 2 and
Year 3 are shown in Table 3.21.
Table 3.21
Product
Number bought in Year 2
Number bought in Year 3
A
B
C
62
44
134
96
46
102
(a) Calculate a current-weighted index (Paasche) for these data using Year 1 as the base
year.
(b) Compare the values with those obtained in Q9.
Section 3.2
Compound interest
Objectives
At the end of this section you should be able to:
●
Understand the difference between simple and compound interest.
●
Calculate the future value of a principal under annual compounding.
●
Calculate the future value of a principal under continuous compounding.
●
Determine the annual percentage rate of interest given a nominal rate of interest.
Today, businesses and individuals are faced with a bewildering array of loan facilities and
investment opportunities. In this section we explain how these financial calculations are car­
ried out to enable an informed choice to be made between the various possibilities available.
We begin by considering what happens when a single lump sum is invested and show how to
calculate the amount accumulated over a period of time.
Suppose that someone gives you the option of receiving $500 now or $500 in three years’
time. Which of these alternatives would you accept? Most people would take the money now,
partly because they may have an immediate need for it, but also because they recognise that
$500 is worth more today than in three years’ time. Even if we ignore the effects of inflation,
it is still better to take the money now, since it can be invested and will increase in value over
the three-year period. In order to work out this value, we need to know the rate of interest
and the basis on which it is calculated. Let us begin by assuming that the $500 is invested for
three years at 10% interest compounded annually. What exactly do we mean by ‘10% inter­
est compounded annually’? Well, at the end of each year, the interest is calculated and is
added on to the amount currently invested. If the original amount is $500, then after one year
the interest is 10% of $500, which is
10
1
3 $500 5
3 $500 5 $50
100
10
so the amount rises by $50 to $550.
What happens to this amount at the end of the second year? Is the interest also $50? This
would actually be the case with simple interest, when the amount of interest received is the
same for all years. However, with compound interest, we get ‘interest on the interest’. Nearly
all financial investments use compound rather than simple interest, because investors need to
be rewarded for not taking the interest payment out of the fund each year. Under annual
compounding the interest obtained at the end of the second year is 10% of the amount
invested at the start of that year. This not only consists of the original $500, but also the $50
already received as interest on the first year’s investment. Consequently, we get an additional
1
3 $550 5 $55
10
Section 3.2 Compound interest 217
raising the sum to $605. Finally, at the end of the third year, the interest is
1
3 $605 5 $60.50
10
so the investment is $665.50. You are therefore $165.50 better off by taking the $500 now and
investing it for three years. The calculations are summarised in Table 3.22.
Table 3.22
End of year
Interest ($)
Investment ($)
1
2
3
50
55
60.50
550
605
665.50
The calculations in Table 3.22 were performed by finding the interest earned each year and
adding it on to the amount accumulated at the beginning of the year. This approach is rather
laborious, particularly if the money is invested over a long period of time. What is really
needed is a method of calculating the investment after, say, 10 years without having to deter­
mine the amount for the nine intermediate years. This can be done using the scale factor
approach discussed in the previous section. To illustrate this, let us return to the problem of
investing $500 at 10% interest compounded annually. The original sum of money is called
the principal and is denoted by P, and the final sum is called the future value and is denoted by
S. The scale factor associated with an increase of 10% is
11
10
5 1.1
100
so at the end of one year the total amount invested is P(1.1).
After two years we get
P(1.1) × (1.1) = P(1.1)2
and after three years the future value is
S = P(1.1)2 × (1.1) = P(1.1)3
Setting P = 500, we see that
S = 500(1.1)3 = $665.50
which is, of course, the same as the amount calculated previously.
In general, if the interest rate is r% compounded annually, then the scale factor is
11
r
100
so, after n years,
S 5 P 11
r
100
n
Given the values of r, P and n, it is simple to evaluate S using the power key x y on a calculator.
218 Chapter 3 Mathematics of Finance
Example
Find the value, in four years’ time, of $10 000 invested at 5% interest compounded annually.
Solution
r
In this problem, P = 10 000, r = 5 and n = 4, so the formula S 5 P 1 1
100
S 5 10 000 1 1
5
100
n
gives
4
5 10 000(1.05)4 5 $12 155.06
Practice Problem
1. Use the formula
r
S 5 P 11
100
n
to find the value, in 10 years’ time, of $1000 invested at 8% interest compounded
annually.
The compound interest formula derived previously involves four variables, r, n, P and S.
Provided that we know any three of these, we can use the formula to determine the remaining
variable. This is illustrated in the following example.
Example
A principal of $25 000 is invested at 12% interest compounded annually. After how many
years will the investment first exceed $250 000?
Solution
We want to save a total of $250 000 starting with an initial investment of $25 000. The
problem is to determine the number of years required for this on the assumption that the
interest is fixed at 12% throughout this time. The formula for compound interest is
r
S 5 P 11
100
n
We are given that
P = 25 000,
S = 250 000,
r = 12
Section 3.2 Compound interest 219
so we need to solve the equation
250 000 5 25 000 1 1
12
100
n
for n.
One way of doing this would just be to keep on guessing values of n until we find the
one that works. However, a more mathematical approach is to use logarithms, because we
are being asked to solve an equation in which the unknown occurs as a power. Following
the method described in Section 2.3, we first divide both sides by 25 000 to get
10 = (1.12)n
Taking logarithms of both sides gives
log(10) = log(1.12)n
and if you apply rule 3 of logarithms you get
log(10) = n log(1.12)
logbxm 5 m logbx
Hence
n5
5
log(10)
log(1.12)
1
(taking logarithms to base 10)
0.49 218 023
(to one decimal place)
5 20.3
Now we know that n must be a whole number because interest is only added on at the end
of each year. We assume that the first interest payment occurs exactly 12 months after the
initial investment and every 12 months thereafter. The answer, 20.3, tells us that after only
20 years the amount is less than $250 000, so we need to wait until 21 years have elapsed
before it exceeds this amount. In fact, after 20 years
S = $25 000(1.12)20 = $241 157.33
and after 21 years
S = $25 000(1.12)21 = $270 096.21
In this example we calculated the time taken for $25 000 to increase by a factor of 10. It
can be shown that this time depends only on the interest rate and not on the actual value of
the principal. To see this, note that if a general principal, P, increases tenfold, then its future
value is 10P. If the interest rate is 12%, then we need to solve
10P 5 P 1 1
12
100
n
for n. The Ps cancel (indicating that the answer is independent of P) to produce the equation
10 = (1.12)n
This is identical to the equation obtained in the previous example and, as we have just seen,
has the solution n = 20.3.
220 Chapter 3 Mathematics of Finance
Practice Problem
2. A firm estimates that its sales will rise by 3% each year and that it needs to sell at least
10 000 goods each year in order to make a profit. Given that its current annual sales
are only 9000, how many years will it take before the firm breaks even?
You may have noticed that in all of the previous problems, it is assumed that the interest is
compounded annually. It is possible for interest to be added to the investment more frequently
than this. For example, suppose that a principal of $500 is invested for three years at 10%
interest compounded quarterly. What do we mean by ‘10% interest compounded quarterly’?
Well, it does not mean that we get 10% interest every three months. Instead, the 10% is split
into four equal portions, one for each quarter. Every three months the interest accrued is
10%
5 2.5%
4
so after the first quarter the investment gets multiplied by 1.025 to give
500(1.025)
and after the second quarter it gets multiplied by another 1.025 to give
500(1.025)2
and so on. Moreover, since there are exactly twelve 3-month periods in three years, we
deduce that the future value is
500(1.025)12 = $672.44
Notice that this is greater than the sum obtained at the start of this section under annual
compounding.
This example highlights the fact that the compound interest formula
S 5 P 11
r
100
n
derived earlier for annual compounding can also be used for other types of compounding. All
that is needed is to reinterpret the symbols r and n. The variable r now represents the rate of
interest per time period, and n represents the total number of periods.
Example
A principal of $10 is invested at 12% interest for one year. Determine the future value if
the interest is compounded
(a) annually (b) semi-annually (c) quarterly (d) monthly (e) weekly
Solution
The formula for compound interest gives
r
S 5 P 11
100
n
Section 3.2 Compound interest 221
(a) If the interest is compounded annually, then r = 12, n = 1, so
S = $10(1.12)1 = $11.20
(b) If the interest is compounded semi-annually, then an interest of 12/2 = 6% is added on
every six months and, since there are two 6-month periods in a year,
S = $10(1.06)2 = $11.24
(c) If the interest is compounded quarterly, then an interest of 12/4 = 3% is added on every
3 months and, since there are four 3-month periods in a year,
S = $10(1.03)4 = $11.26
(d) If the interest is compounded monthly, then an interest of 12/12 = 1% is added on every
month and, since there are 12 months in a year,
S = $10(1.01)12 = $11.27
(e) If the interest is compounded weekly then an interest of 12/52 = 0.23% is added on
every week and, since there are 52 weeks in a year,
S = $10(1.0023)52 = $11.27
In the above example we see that the future value rises as the frequency of compounding
rises. This is to be expected because the basic feature of compound interest is that we get
‘interest on the interest’. However, one important observation that you might not have
expected is that, although the future values increase, they appear to be approaching a fixed
value. It can be shown that this always occurs. The type of compounding in which the inter­
est is added on with increasing frequency is called continuous compounding. In theory, we can
find the future value of a principal under continuous compounding using the approach taken
in the previous example. We work with smaller and smaller time periods until the numbers
settle down to a fixed value. However, it turns out that there is a special formula that can be
used to compute this directly. The future value, S, of a principal, P, compounded continu­
ously for t years at an annual rate of r% is
S = Pert/100
where e is the number
2.718 281 828 459 045 235 36 (to 20 decimal places)
If r = 12, t = 1 and P = 10, then this formula gives
S = $10e12×1/100 = $10e0.12 = $11.27
check this using
your own
calculator
which is in agreement with the limiting value obtained in the previous example.
222 Chapter 3 Mathematics of Finance
Advice
The number e and the related natural logarithm function were first introduced in Section 2.4. If you
missed this section, you should go back and read through this work now before proceeding. The
link between the number e and the above formula for continuous compounding is given in Question
7 in Exercise 3.2* at the end of this section. However, you may prefer to accept it without justification
and concentrate on the applications.
Example
A principal of $2000 is invested at 10% interest compounded continuously. After how
many days will the investment first exceed $2100?
Solution
We want to save a total of $2100 starting with an initial investment of $2000. The problem
is to determine the number of days required for this on the assumption that the interest
rate is 10% compounded continuously. The formula for continuous compounding is
S = Pert/100
We are given that
S = 2100, P = 2000, r = 10
so we need to solve the equation
2100 = 2000e10t/100
for t. Dividing through by 2000 gives
1.05 = e0.1t
As explained in Section 2.4, equations such as this can be solved using natural loga­
rithms. Recall that
if M = en then
n = ln M
If we apply this definition to the equation
1.05 = e0.1t
with M = 1.05 and n = 0.1t, then
0.1t = ln(1.05) = 0.048 790 2
and so t = 0.488 to three decimal places.
The variable t which appears in the formula for continuous compounding is measured
in years, so to convert it to days we multiply by 365 (assuming that there are 365 days in
a year). Hence
t = 365 × 0.488 = 178.1 days
We deduce that the amount invested first exceeds $2100 some time during the 179th day.
Section 3.2 Compound interest 223
Practice Problems
3. (1)A principal, $30, is invested at 6% interest for two years. Determine the future
value if the interest is compounded
(a) annually
(b) semi-annually
(c) quarterly
(d) monthly
(e) weekly
(f) daily
(2) Use the formula
S = Pert/100
to determine the future value of $30 invested at 6% interest compounded continu­
ously for two years. Confirm that it is in agreement with the results of part (1).
4. Determine the rate of interest required for a principal of $1000 to produce a future
value of $4000 after 10 years compounded continuously.
Given that there are so many ways of calculating compound interest, people often find it difficult
to appraise different investment opportunities. What is needed is a standard ‘benchmark’ that
enables an individual to compare different forms of savings or credit schemes on an equal
basis. The one that is commonly used is annual compounding. All firms offering investment
or loan facilities are required to provide the effective annual rate. This is often referred to as
the annual percentage rate, which is abbreviated to APR. The APR is the rate of interest which,
when compounded annually, produces the same yield as the nominal (that is, the stated) rate of
interest. The phrase ‘annual equivalent rate’ (AER) is frequently used when applied to savings.
Example
Determine the annual equivalent rate of interest of a deposit account that has a nominal
rate of 6.6% compounded monthly.
Solution
The AER is the overall rate of interest, which can be calculated using scale factors. If the
account offers a return of 6.6% compounded monthly, then each month the interest is
6.6
5 0.55%
12
of the amount invested at the beginning of that month. The monthly scale factor is
11
0.55
5 1.0055
100
so in a whole year the principal gets multiplied by
(1.0055)12 = 1.068
which can be written as
11
6.8
100
so the AER is 6.8%.
224 Chapter 3 Mathematics of Finance
Practice Problem
5. Determine the annual percentage rate of interest if the nominal rate is 12% com­
pounded quarterly.
Although the aim of this chapter is to investigate the mathematics of finance, the math­
ematical techniques themselves are more widely applicable. We conclude this section with
two examples to illustrate this.
Example
A country’s annual GNP (gross national product), currently at $25 000 million, is pre­
dicted to grow by 3.5% each year. The population is expected to increase by 2% a year
from its current level of 40 million. After how many years will GNP per capita (that is,
GNP per head of population) reach $700?
Solution
The per capita value of GNP is worked out by dividing GNP by the size of the population.
Initially, this is
25 000 000 000
5 $625
40 000 000
During the next few years, GNP is forecast to grow at a faster rate than the population,
so this value will increase.
The scale factor associated with a 3.5% increase is 1.035, so after n years GNP (in mil­
lions of dollars) will be
GNP = 25 000 × (1.035)n
Similarly, the population (also in millions) will be
population = 40 × (1.02)n
Hence GNP per capita is
(1.035)n
25 000 3 (1.035)n
25 000
1.035
5
3
5 625 3
n
n
40 3 (1.02)
(1.02)
40
1.02
n
We want to find the number of years required for this to reach 700, so we need to solve the
equation
625 3
1.035
1.02
n
5 700
for n. Dividing both sides by 625 gives
1.035
1.02
n
5 1.12
Section 3.2 Compound interest 225
and after taking logs of both sides we get
log
n
1.035
1.02
5 log(1.12)
1.035
nlog
1.02
n
5 log(1.12) (rule 3 of logs)
so that
n5
log(1.12)
5 7.76
log(1.035/1.02)
We deduce that the target figure of $700 per capita will be achieved after eight years.
Example
A firm decides to increase output at a constant rate from its current level of 50 000 to 60 000
during the next five years. Calculate the annual rate of increase required to achieve this growth.
Solution
If the rate of increase is r%, then the scale factor is 1 1
r
50 000 1 1
100
r
so, after five years, output will be
100
5
To achieve a final output of 60 000, the value of r is chosen to satisfy the equation
50 000 1 1
r
100
5
5 60 000
Dividing both sides by 50 000 gives
r
11
100
5
5 1.2
The difficulty in solving this equation is that the unknown, r, is trapped inside the brackets,
which are raised to the power of 5. This is analogous to the problem of solving an equation
such as
x2 = 5.23
which we would solve by taking square roots of both sides to find x. This suggests that we
can find r by taking fifth roots of both sides of
r
11
100
5
5 1.2
to get
11
r
5 (1.2)1/5 5 1.037
100
Hence r = 3.7%.
226 Chapter 3 Mathematics of Finance
Practice Problem
6. The turnover of a leading supermarket chain, A, is currently $560 million and is
expected to increase at a constant rate of 1.5% a year. Its nearest rival, supermarket B,
has a current turnover of $480 million and plans to increase this at a constant rate of
3.4% a year. After how many years will supermarket B overtake supermarket A?
Key Terms
Annual percentage rate The equivalent annual interest paid for a loan, taking into account the
compounding over a variety of time periods.
Compound interest The interest which is added on to the initial investment, so that this will itself
gain interest in subsequent time periods.
Continuous compounding The limiting value when interest is compounded with ever-increasing
frequency.
Future value
The final value of an investment after one or more time periods.
Principal The value of the original sum invested.
Simple interest The interest which is paid directly to the investor instead of being added to the
original amount.
Exercise 3.2
1. A bank offers a return of 7% interest compounded annually. Find the future value of
a principal of $4500 after six years. What is the overall percentage rise over this period?
2. Find the future value of $20 000 in two years’ time if compounded quarterly at 8% interest.
3. The value of an asset, currently priced at $100 000, is expected to increase by 20% a
year.
(a) Find its value in 10 years’ time.
(b) After how many years will it be worth $1 million?
4. How long will it take for a sum of money to double if it is invested at 5% interest com­
pounded annually?
5. A piece of machinery depreciates in value by 5% a year. Determine its value in three
years’ time if its current value is $50 000.
6. A principal, $7000, is invested at 9% interest for eight years. Determine its future value
if the interest is compounded
(a) annually (b) semi-annually (c) monthly (d) continuously
Section 3.2 Compound interest 227
7. Which of the following savings accounts offers the greater return?
Account A: an annual rate of 8.05% paid semi-annually
Account B: an annual rate of 7.95% paid monthly
8. Find the future value of $100 compounded continuously at an annual rate of 6% for
12 years.
9. How long will it take for a sum of money to triple in value if invested at an annual rate
of 3% compounded continuously?
10. If a piece of machinery depreciates continuously at an annual rate of 4%, how many
years will it take for the value of the machinery to halve?
11. A department store has its own credit card facilities, for which it charges interest at a
rate of 2% each month. Explain briefly why this is not the same as an annual rate of
24%. What is the annual percentage rate?
12. Determine the APR if the nominal rate is 7% compounded continuously.
13. Current annual consumption of energy is 78 billion units, and this is expected to rise at
a fixed rate of 5.8% each year. The capacity of the industry to supply energy is currently
104 billion units.
(a) Assuming that the supply remains steady, after how many years will demand exceed
supply?
(b) What constant rate of growth of energy production would be needed to satisfy
demand for the next 50 years?
14. Find the value, in two years time, of $4000 invested at 5% compounded annually. In the
following two years, the interest rate is expected to rise to 8%. Find the final value of
the investment at the end of the four-year period, and find the overall percentage
increase. Give your answers correct to two decimal places.
15. Find the APR of a loan if the monthly interest rate is 1.65%. Give your answer correct
to two decimal places.
16. The future value S of principal P invested for n years with an interest rate r% com­
pounded annually may be calculated using the formula
r
S 5 P 11
100
n
Rearrange this formula to express P in terms of S, r and n.
17. The number of rail passenger journeys made between England and Scotland in 2004
was 5.015 million. In 2011 the figure was 7.419 million. Work out the yearly percentage
rate of growth (assumed constant) during this period.
18. Table 3.23 shows the depreciation in the value of two models of car.
Table 3.23
Year
Year 1
Year 2
Car A
Car B
36 000
32 000
32 000
28 800
(a) Assuming that the depreciation of Car A is linear, estimate its value in Year 3.
(b) Assuming that the depreciation of Car B is exponential, estimate its value in Year 3.
(c) Predict when Car A will be worth less than Car B.
228 Chapter 3 Mathematics of Finance
Exercise 3.2*
1. A principal of $7650 is invested at a rate of 3.7% compounded annually. After how
many years will the investment first exceed $12 250?
2. A principal of $70 000 is invested at 6% interest for four years. Find the difference in
the future value if the interest is compounded quarterly compared to continuous com­
pounding. Round your answer to two decimal places.
3. Midwest Bank offers a return of 5% compounded annually for each and every year. The
rival BFB offers a return of 3% for the first year and 7% in the second and subsequent
years (both compounded annually). Which bank would you choose to invest in if you
decided to invest a principal for (a) two years; (b) three years?
4. A car depreciates by 40% in the first year, 30% in the second year and 20% thereafter.
I buy a car for $14 700 when it is two years old.
(a) How much did it cost when new?
(b) After how many years will it be worth less than 25% of the amount that I paid for it?
5. The population of a country is currently at 56 million and is forecast to rise by 3.7%
each year. It is capable of producing 2500 million units of food each year, and it is
estimated that each member of the population requires a minimum of 65 units of food
each year. At the moment, the extra food needed to satisfy this requirement is imported,
but the government decides to increase food production at a constant rate each year,
with the aim of making the country self-sufficient after 10 years. Find the annual rate
of growth required to achieve this.
6. Simon decides to buy a new sofa which is available at each of three stores at the same
fixed price. He decides to borrow the money using each store’s credit facility.
Store A has an effective rate of interest of 12.6%.
Store B charges interest at a rate of 10.5% compounded continuously.
Store C charges interest at a rate of 11.5% compounded quarterly.
From which store should Simon buy his sofa to minimise the total cost?
7. If a principal, P, is invested at r% interest compounded annually, then its future value,
S, after n years is given by
S 5 P 11
r
100
n
(a) Use this formula to show that if an interest rate of r% is compounded k times a year,
then after t years
tk
r
S 5 P 11
100k
(b) Show that if m = 100k/r, then the formula in part (a) can be written as
S5P
1
11
m
m
rt /100
Section 3.2 Compound interest 229
(c) Use the definition
e 5 lim 1 1
m
`
1
m
m
to deduce that if the interest is compounded with ever-increasing frequency (that is,
continuously), then
S = Pert/100
8. World oil reserves are currently estimated to be 600 billion units. If this quantity is
reduced by 8% a year, after how many years will oil reserves drop below 100 billion
units?
9. The nominal rate of interest of a store card is 18% compounded monthly.
(a) State the monthly interest rate.
(b) Find the equivalent annual rate of interest if the compounding is continuous. Round
your answer to two decimal places.
10. Write down an expression for the annual percentage growth rate of a country whose
GDP increases by a factor of g over a period of n years.
11. (a) A principal, P, is invested at a nominal rate of interest, of r% compounded n times
a year. Show that the AER is given by the formula
r
AER 5 100 1 1
100n
n
2 100
(b) Find a formula for AER when the interest is compounded continuously.
Section 3.3
Geometric series
Objectives
At the end of this section you should be able to:
●
Recognise a geometric progression.
●
Evaluate a geometric series.
●
Calculate the total investment obtained from a regular savings plan.
●
Calculate the instalments needed to repay a loan.
Consider the following sequence of numbers:
2, 6, 18, 54, . . . One obvious question, often asked in intelligence tests, is, what is the next term in the
sequence? All that is required is for you to spot the pattern so that it can be used to generate
the next term. In this case, successive numbers are obtained by multiplying by 3, so the fifth
term is
54 × 3 = 162
the sixth term is
162 × 3 = 486
and so on. Any sequence in which terms are calculated by multiplying their predecessor by a
fixed number is called a geometric progression, and the multiplicative factor itself is called
a geometric ratio. The sequence above is a geometric progression with geometric ratio 3. The
reason for introducing these sequences is not to help you to answer intelligence tests but
rather to analyse compound interest problems. You may well have noticed that all of the
problems given in the previous section produced such a sequence. For example, if a principal,
$500, is invested at 10% interest compounded annually, then the future values in successive
years are
500(1.1), 500(1.1)2, 500(1.1)3, . . .
which we recognise as a geometric progression with geometric ratio 1.1.
Example
Which of the following sequences are geometric progressions? For those sequences that are
of this type, write down their geometric ratios.
(a) 1000, −100, 10, −1, . . . (b) 2, 4, 6, 8, . . . (c) a, ar, ar2, ar3, . . .
Section 3.3 Geometric series 231
Solution
(a) 1000, −100, 10, −1, . . . is a geometric progression with geometric ratio 2
1
.
10
(b) 2, 4, 6, 8, . . . is not a geometric progression because to go from one term to the next,
you add 2. Such a sequence is called an arithmetic progression and is of little interest in
business and economics.
(c) a, ar, ar 2, ar 3, . . . is a geometric progression with geometric ratio r.
Practice Problem
1. Decide which of the following sequences are geometric progressions. For those
sequences that are of this type, write down their geometric ratios.
(a) 3, 6, 12, 24, . . .
(b) 5, 10, 15, 20, . . . (c) 1, −3, 9, −27, . . .
(d) 8, 4, 2, 1, 1/2, . . .
(e) 500, 500(1.07), 500(1.07)2, . . .
All of the problems considered in Section 3.2 involved a single lump-sum payment into an
investment account. The task was simply to determine its future value after a period of time
when it is subject to a certain type of compounding. In this section, we extend this to include
multiple payments. This situation occurs whenever individuals save regularly or when busi­
nesses take out a loan that is paid back using fixed monthly or annual instalments. To tackle
these problems we need to be able to sum (that is, to add together) consecutive terms of a
geometric progression. Such an expression is called a geometric series. Suppose that we want
to sum the first six terms of the geometric progression given by the sequence
2, 6, 18, 54, . . . (1)
The easiest way of doing this is to write down these six numbers and add them together to get
2 + 6 + 18 + 54 + 162 + 486 = 728
There is, however, a special formula to sum a geometric series which is particularly useful
when there are lots of terms or when the individual terms are more complicated to evaluate.
It can be shown that the sum of the first n terms of a geometric progression in which the first
term is a, and the geometric ratio is r, is equal to
a
rn 2 1
r21
(r Þ 1)
Use of the symbol r to denote both the interest rate and the geometric ratio is unfortunate but
fairly standard. In practice, it is usually clear from the context what this symbol represents,
so no confusion should arise.
232 Chapter 3 Mathematics of Finance
A proof of this formula is given in Question 7 in Exercise 3.3* at the end of this section.
As a check, let us use it to determine the sum of the first six terms of sequence (1) considered
previously. In this case the first term a = 2, the geometric ratio r = 3 and the number of terms
n = 6, so the geometric series is equal to
2
36 2 1
5 36 2 1 5 728
321
which agrees with the previous value found by summing the terms longhand. In this case
there is no real benefit in using the formula. However, it would be tedious to evaluate the
geometric series
500(1.1) + 500(1.1)2 + 500(1.1)3 + . . . + 500(1.1)25
longhand, whereas substituting a = 500(1.1), r = 1.1 and n = 25 into the formula immedi­
ately gives
500(1.1)
(1.1)25 2 1
5 54 090.88
1.1 2 1
Practice Problem
2. (a) Write down the next term in the sequence
1, 2, 4, 8, . . . and hence find the sum of the first five terms. Check that this agrees with the value
obtained using
a
rn 2 1
r21
(b) Evaluate the geometric series
100(1.07) + 100(1.07)2 + . . . + 100(1.07)20
There are two particular applications of geometric series that we now consider, involving
savings and loans. We begin by analysing savings plans. In the simplest case, an individual
decides to invest a regular sum of money into a bank account. This is sometimes referred to
as a sinking fund and is used to meet some future financial commitment. It is assumed that he
or she saves an equal amount and that the money is put into the account at the same time
each year (or month). We further assume that the interest rate does not change. The latter
may not be an entirely realistic assumption, since it can fluctuate wildly in volatile market
conditions. Indeed, banks offer a variety of rates of interest depending on the notice required
for withdrawal and on the actual amount of money saved. Question 5 in Exercise 3.3* at the
end of this section considers what happens when the interest rate rises as the investment goes
above certain threshold levels.
Section 3.3 Geometric series 233
Example
A person saves $100 in a bank account at the beginning of each month. The bank offers a
return of 12% compounded monthly.
(a) Determine the total amount saved after 12 months.
(b) After how many months does the amount saved first exceed $2000?
Solution
(a) During the year a total of 12 regular savings of $100 are made. Each $100 is put into an
account that gives a return of 12% compounded monthly, or equivalently, a return of
1% each month. However, each payment is invested for a different period of time. For
example, the first payment is invested for the full 12 months, whereas the final payment
is invested for 1 month only. We need to work out the future value of each payment
separately and add them together.
The first payment is invested for 12 months, gaining a monthly interest of 1%, so its
future value is
100(1.01)12
The second payment is invested for 11 months, so its future value is
100(1.01)11
Likewise, the third payment yields
100(1.01)10
and so on. The last payment is invested for 1 month, so its future value is
100(1.01)1
The total value of the savings at the end of 12 months is then
100(1.01)12 + 100(1.01)11 + . . . + 100(1.01)1
If we rewrite this series in the order of ascending powers, we then have the more familiar
form
100(1.01)1 + . . . + 100(1.01)11 + 100(1.01)12
This is equal to the sum of the first 12 terms of a geometric progression in which the first
term is 100(1.01) and the geometric ratio is 1.01. Its value can therefore be found by using
a
rn 2 1
r21
with a = 100(1.01), r = 1.01 and n = 12, which gives
$100(1.01)
(1.01)12 2 1
5 $1280.93
1.01 2 1
(b) In part (a) we showed that after 12 months the total amount saved is
100(1.01) + 100(1.01)2 + . . . + 100(1.01)12
Using exactly the same argument, it is easy to see that after n months the account contains
100(1.01) + 100(1.01)2 + . . . + 100(1.01)n
➜
234 Chapter 3 Mathematics of Finance
The formula for the sum of the first n terms of a geometric progression shows that this
is the same as
100(1.01)
1.01n 2 1
5 10 100(1.01n 2 1)
1.01 2 1
The problem here is to find the number of months needed for total savings to rise to
$2000. Mathematically, this is equivalent to solving the equation
10 100(1.01n − 1) = 2000
for n. Following the strategy described in Section 2.3 gives
1.01n 2 1 5 0.198
n
1.01 5 1.198
(divide both sides by 10 100)
(add 1 to both sides)
log(1.01)n 5 log(1.198)
(take logs of both sides)
nlog(1.01) 5 log(1.198)
(rule 3 of logs)
n5
log(1.198)
(divide both sides by log(1.01))
log(1.01)
5 18.2
It follows that after 18 months savings are less than $2000, whereas after 19 months
savings exceed this amount. The target figure of $2000 is therefore reached at the end
of the 19th month.
Practice Problem
3. An individual saves $1000 in a bank account at the beginning of each year. The bank
offers a return of 8% compounded annually.
(a) Determine the amount saved after 10 years.
(b) After how many years does the amount saved first exceed $20 000?
We now turn our attention to loans. Many businesses finance their expansion by obtaining
loans from a bank or other financial institution. Banks are keen to do this provided that
they receive interest as a reward for lending money. Businesses pay back loans by monthly
or annual repayments. The way in which this repayment is calculated is as follows. Let us
suppose that interest is calculated on a monthly basis and that the firm repays the debt by
fixed monthly instalments at the end of each month. The bank calculates the interest charged
during the first month based on the original loan. At the end of the month, this interest is
added on to the original loan, and the repayment is simultaneously deducted to determine
the amount owed. The bank then charges interest in the second month based on this new
amount, and the process is repeated. Provided that the monthly repayment is greater than the
interest charged each month, the amount owed decreases and, eventually, the debt is cleared.
In practice, the period during which the loan is repaid is fixed in advance and the monthly
repayments are calculated to achieve this end.
Section 3.3 Geometric series 235
Example
Determine the monthly repayments needed to repay a $100 000 loan which is paid back
over 25 years when the interest rate is 8% compounded annually.
Solution
In this example the time interval between consecutive repayments is 1 month, whereas
the period during which interest is charged is one year. This type of financial calculation
typifies the way in which certain types of housing loan are worked out. The interest is
compounded annually at 8%, so the amount of interest charged during the first year is 8%
of the original loan: that is,
8
3 100 000 5 8000
100
This amount is added on to the outstanding debt at the end of the first year. During this
time, 12 monthly repayments are made, so if each instalment is $x, the outstanding debt
must decrease by 12x. Hence, at the end of the first year, the amount owed is
100 000 + 8000 − 12x = 108 000 − 12x
In order to be able to spot a pattern in the annual debt, let us write this as
100 000(1.08) − 12x
where the first part simply reflects the fact that 8% interest is added on to the original sum
of $100 000. At the end of the second year, a similar calculation is performed. The amount
owed rises by 8% to become
[100 000(1.08) − 12x](1.08) = 100 000(1.08)2 − 12x(1.08)
and we deduct 12x for the repayments to get
100 000(1.08)2 − 12x(1.08) − 12x
This is the amount owed at the end of the second year. Each year we multiply by 1.08 and
subtract 12x, so at the end of the third year we owe
[100 000(1.08)2 − 12x(1.08) − 12x] (1.08) − 12x
= 100 000(1.08)3 − 12x(1.08)2 − 12x(1.08) − 12x
and so on. These results are summarised in Table 3.24. If we continue the pattern, we see
that after 25 years the amount owed is
100 000(1.08)25 − 12x(1.08)24 − 12x(1.08)23 − . . . − 12x
= 100 000(1.08)25 − 12x[1 + 1.08 + (1.08)2 + . . . + (1.08)24]
(Taking out a common factor of 12x and rewriting powers of 1.08 in ascending order.)
Table 3.24
End of year
Outstanding debt
1
2
3
100 000(1.08)1 − 12x
100 000(1.08)2 − 12x(1.08)1 − 12x
100 000(1.08)3 − 12x(1.08)2 − 12x(1.08)1 − 12x
➜
236 Chapter 3 Mathematics of Finance
The first term is easily evaluated using a calculator to get
100 000(1.08)25 = 684 847.520
The geometric series inside the square brackets can be worked out from the formula
a
rn 2 1
r21
The first term a = 1, the geometric ratio r = 1.08, and we are summing the first 25 terms,
so n = 25. (Can you see why there are actually 25 terms in this series rather than 24?) Hence
[1 1 1.08 1 (1.08)2 1 . . . 1 (1.08)24] 5
1.0825 2 1
5 73.106
1.08 2 1
The amount owed at the end of 25 years is therefore
684 847.520 − 12x(73.106) = 684 847.520 − 877.272x
In this expression, x denotes the monthly repayment, which is chosen so that the debt is
completely cleared after 25 years. This will be so if x is the solution of
684 847.520 − 877.272x = 0
Hence
x5
684 847.520
5 $780.66
877.272
The monthly repayment on a 25-year loan of $100 000 is $780.66, assuming that the interest
rate remains fixed at 8% throughout this period.
It is interesting to substitute this value of x into the expressions for the outstanding debt
given in Table 3.24. The results are listed in Table 3.25. What is so depressing about these
figures is that the debt falls by only about $1500 to begin with, in spite of the fact that over
$9000 is being repaid each year!
Table 3.25
End of year
Outstanding debt
1
2
3
$98 632.08
$97 154.73
$95 559.18
Practice Problem
4. A person requests an immediate bank overdraft of $2000. The bank generously agrees
to this but insists that it should be repaid by 12 monthly instalments and charges 1%
interest every month on the outstanding debt. Determine the monthly repayment.
The mathematics used in this section for problems on savings and loans can be used for
other time series. Reserves of non-renewable commodities such as minerals, oil and gas con­
tinue to decline, and geometric series can be used to estimate the year in which these stocks
are likely to run out.
Section 3.3 Geometric series 237
Example
Total reserves of a non-renewable resource are 250 million tonnes. Annual consumption,
currently at 20 million tonnes per year, is expected to rise by 2% a year. After how many
years will stocks be exhausted?
Solution
In the first year, consumption will be 20 million tonnes. In the second year, this will rise
by 2%, so consumption will be 20(1.02) million tonnes. In the third year, this will again
rise by 2% to become 20(1.02)2 million tonnes. The total consumption (in millions of
tonnes) during the next n years will be
20 + 20(1.02) + 20(1.02)2 + . . . + 20(1.02)n−1
This represents the sum of n terms of a geometric series with first term a = 20 and geometric
ratio r = 1.02, so is equal to
20
1.02n 2 1
5 1000(1.02n 2 1)
1.02 2 1
Reserves will run out when this exceeds 250 million, so we need to solve the equation
1000(1.02n − 1) = 250
for n. This is easily solved using logarithms:
1.02n 2 1 5 0.25
(divide both sides by 1000)
n
(add 1 to both sides)
n
log(1.02) 5 log(1.25)
(take logs of both sides)
nlog(1.02) 5 log(1.25)
(rule 3 of logs)
1.02 5 1.25
log(1.25)
n5
log(1.02)
(divide both sides by log(1.02))
5 11.27
so the reserves will be completely exhausted after 12 years.
Practice Problem
5. It is estimated that world reserves of oil currently stand at 2625 billion units. Oil is
currently extracted at an annual rate of 45.5 billion units and this is set to increase by
2.6% a year. After how many years will oil reserves run out?
Example
Current annual extraction of a non-renewable resource is 40 billion units, and this is
expected to fall at a rate of 5% each year. Estimate the current minimum level of reserves
if this resource is to last in perpetuity (that is, for ever).
➜
238 Chapter 3 Mathematics of Finance
Solution
In the first year 40 billion units are extracted. In the second year this falls by 5% to 40(0.95)
billion units. In the third year this goes down by a further 5% to 40(0.95)2. After n years
the total amount extracted will be
40 + 40(0.95) + 40(0.95)2 + . . . + 40(0.95)n−1
Using the formula for the sum of a geometric progression gives
40
0.95n 2 1
0.95n 2 1
5 40
5 800(1 2 0.95n)
0.95 2 1
20.05
To see what happens in perpetuity we need to investigate the behaviour 0.95n as n tends
to infinity. Since the magnitude of 0.95 is less than unity, it is easy to see that 0.95n con­
verges to zero and so the total amount will be 800 billion units.
Key Terms
Arithmetic progression A sequence of numbers with a constant difference between consecutive
terms; the nth term takes the form, a + bn.
Geometric progression A sequence of numbers with a constant ratio between consecutive
terms; the nth term takes the form, ar n−1.
Geometric ratio
The constant multiplier in a geometric series.
Geometric series A sum of the consecutive terms of a geometric progression.
Sinking fund A fixed sum of money saved at regular intervals which is used to fund some future
financial commitment.
Exercise 3.3
1. Find the value of the geometric series
1000 + 1000(1.03) + 1000(1.03)2 + . . . + 1000(1.03)9
2. An individual saves $5000 in a bank account at the beginning of each year for 10 years.
No further savings or withdrawals are made from the account. Determine the total
amount saved if the annual interest rate is 8% compounded:
(a) annually;
(b) semi-annually.
3. Determine the monthly repayments needed to repay a $125 000 loan which is paid back
over 20 years when the interest rate is 7% compounded annually. Round your answer to
two decimal places.
Section 3.3 Geometric series 239
4. A prize fund is set up with a single investment of $5000 to provide an annual prize of
$500. The fund is invested to earn interest at a rate of 7% compounded annually. If the
first prize is awarded one year after the initial investment, find the number of years for
which the prize can be awarded before the fund falls below $500.
5. The current extraction of a certain mineral is 12 million tonnes a year, and this is
expected to fall at a constant rate of 6% each year. Estimate the current minimum level
of world reserves if the extraction is to last in perpetuity.
6. A person invests $5000 at the beginning of a year in a savings account that offers a
return of 4.5% compounded annually. At the beginning of each subsequent year an
additional $1000 is invested in the account. How much will there be in the account at
the end of 10 years?
7. A person borrows $100 000 at the beginning of a year and agrees to repay the loan in
ten equal instalments at the end of each year. Interest is charged at a rate of 6% com­
pounded annually.
(a) Find the annual repayment.
(b) Work out the total amount of interest paid and compare this with the total interest
paid when repaying the loan in five equal annual instalments instead of ten.
8. A person wishes to save a regular amount at the beginning of each month in order to
buy a car in 18 months’ time. An account offers a return of 4.8% compounded monthly.
Work out the monthly savings if the total amount saved at the end of 18 months is
$18 000.
Exercise 3.3*
1. Find the sum of the of the geometric series,
5 − 20 + 80 − 320 + . . . − 20 971 520
2. A regular saving of $500 is made into a sinking fund at the start of each year for
10 years. Determine the value of the fund at the end of the 10th year on the assumption
that the rate of interest is
(a) 11% compounded annually;
(b) 10% compounded continuously.
3. Monthly sales figures for January are 5600. This is expected to fall for the following
9 months at a rate of 2% each month. Thereafter sales are predicted to rise at a constant
rate of 4% each month. Estimate total sales for the next two years (including the first
January).
4. Determine the monthly repayments needed to repay a $50 000 loan that is paid back
over 25 years when the interest rate is 9% compounded annually. Calculate the increased
monthly repayments needed in the case when
(a) the interest rate rises to 10%;
(b) the period of repayment is reduced to 20 years.
➜
240 Chapter 3 Mathematics of Finance
5. A bank has three different types of account in which the interest rate depends on the
amount invested. The ‘ordinary’ account offers a return of 6% and is available to every
customer. The ‘extra’ account offers 7% and is available only to customers with $5000
or more to invest. The ‘superextra’ account offers 8% and is available only to customers
with $20 000 or more to invest. In each case, interest is compounded annually and is
added to the investment at the end of the year.
A person saves $4000 at the beginning of each year for 25 years. Calculate the total
amount saved at the end of 25 years on the assumption that the money is transferred to
a higher-interest account at the earliest opportunity.
6. A business takes out a loan of $500 000 from a bank and agrees to repay the loan by
paying a fixed amount of $60 000 at the end of each subsequent year. Once the debt
falls below $60 000 the business pays off the outstanding debt as the final payment.
Work out the final payment if the interest rate is 7.5% compounded annually.
7. If
Sn = a + ar + ar2 + . . . + arn−1
write down an expression for rSn and deduce that
rSn − Sn = ar n − a
Hence show that the sum of the first n terms of a geometric progression with first term
a and geometric ratio r is given by
a
rn 2 1
r 21
provided that r ≠ 1.
8. At the beginning of a month, a customer owes a credit card company $8480. In the
middle of the month, the customer repays $A, where A < $8480, and at the end of the
month the company adds interest at a rate of 6% of the outstanding debt. This process
is repeated with the customer continuing to pay off the same amount, $A, each month.
(a) Find the value of A for which the customer still owes $8480 at the start of each
month.
(b) If A = 1000, calculate the amount owing at the end of the eighth month.
(c) Show that the value of A for which the whole amount owing is exactly paid off after
the nth payment is given by
A5
8480Rn21(R 2 1)
Rn 2 1
where R 5 1.06
(d) Find the value of A if the debt is to be paid off exactly after two years.
Section 3.4
Investment appraisal
Objectives
At the end of this section you should be able to:
●
Calculate present values under discrete and continuous compounding.
●
Use net present values to appraise investment projects.
●
Calculate the internal rate of return.
●
Calculate the present value of an annuity.
●
Use discounting to compare investment projects.
●
Calculate the present value of government securities.
In Section 3.2 the following two formulas were used to solve compound interest problems
S 5 P 11
t
r
100
(1)
S = Pert/100
(2)
The first of these can be applied to any type of compounding in which the interest is added
on to the investment at the end of discrete time intervals. The second formula is used when
the interest is added on continuously. Both formulas involve the variables
P = principal
S = future value
r = interest rate
t = time
In the case of discrete compounding, the letter t represents the number of time periods. (In
Section 3.2 this was denoted by n.) For continuous compounding, t is measured in years.
Given any three of these variables, it is possible to work out the value of the remaining vari­
able. Various examples were considered in Section 3.2. Of particular interest is the case
where S, r and t are given, and P is the unknown to be determined. In this situation we know
the future value, and we want to work backwards to calculate the original principal. This
process is called discounting and the principal, P, is called the present value. The rate of inter­
est is sometimes referred to as the discount rate. Equations (1) and (2) are easily rearranged to
produce explicit formulas for the present value under discrete and continuous compounding:
S
r
P5
5 S 11
(1 1 r/100)t
100
P5
S
ert/100
2rt/100
5 Se
2t
reciprocals are
denoted by
negative powers
242 Chapter 3 Mathematics of Finance
Example
Find the present value of $1000 in four years’ time if the discount rate is 10% compounded
(a) semi-annually
(b) continuously
Solution
(a) The discount formula for discrete compounding is
P 5 S 11
r
100
2t
If compounding occurs semi-annually, then r = 5 since the interest rate per six months
is 10/2 = 5, and t = 8 since there are eight 6-month periods in four years. We are given
that the future value is $1000, so
P = $1000(1.05)−8 = $676.84
(b) The discount formula for continuous compounding is
P = Se−rt/100
In this formula, r is the annual discount rate, which is 10, and t is measured in years,
so is 4. Hence the present value is
P = $1000e−0.4 = $670.32
Notice that the present value in part (b) is smaller than that in part (a). This is to
be expected because continuous compounding always produces a higher yield.
Consequently, we need to invest a smaller amount under continuous compounding to
produce the future value of $1000 after four years.
Practice Problem
1. Find the present value of $100 000 in 10 years’ time if the discount rate is 6% compounded
(a) annually (b) continuously
Present values are a useful way of appraising investment projects. Suppose that you are
invited to invest $600 today in a business venture that is certain to produce a return of $1000
in five years’ time. If the discount rate is 10% compounded semi-annually, then part (a) of the
pre­vious example shows that the present value of this return is $676.84. This exceeds the initial
outlay of $600, so the venture is regarded as profitable. We quantify this profit by calculating
the difference between the present value of the revenue and the present value of the costs,
which is known as the net present value (NPV). In this example, the net present value is
$676.84 − $600 = $76.84
Quite generally, a project is considered worthwhile when the NPV is positive. Moreover, if
a decision is to be made between two different projects, then the one with the higher NPV is
the preferred choice.
Section 3.4 Investment appraisal 243
An alternative way of assessing individual projects is based on the internal rate of return
(IRR). This is the annual rate which, when applied to the initial outlay, yields the same return
as the project after the same number of years. The investment is considered worthwhile pro­
vided the IRR exceeds the market rate. Obviously, in practice, other factors such as risk need
to be considered before a decision is made.
The following example illustrates both NPV and IRR methods and shows how a value of
the IRR itself can be calculated.
Example
A project requiring an initial outlay of $15 000 is guaranteed to produce a return of
$20 000 in three years’ time. Use the
(a) net present value
(b) internal rate of return
methods to decide whether this investment is worthwhile if the prevailing market rate is
5% compounded annually. Would your decision be affected if the interest rate were 12%?
Solution
(a) The present value of $20 000 in three years’ time, based on a discount rate of 5%, is
found by setting S = 20 000, t = 3 and r = 5 in the formula
r
P 5 S 11
100
2t
This gives
P = $20 000(1.05)−3 = $17 276.75
The NPV is therefore
$17 276.75 − $15 000 = $2276.75
The project is to be recommended because this value is positive.
(b) To calculate the IRR we use the formula
S 5 P 11
t
r
100
We are given S = 20 000, P = 15 000 and t = 3, so we need to solve
r
20 000 5 15 000 1 1
100
3
for r. An obvious first step is to divide both sides of this equation by 15 000 to get
4
r
5 11
3
100
3
➜
244 Chapter 3 Mathematics of Finance
We can extract r by taking cube roots of both sides of
11
r
100
3
5
4
3
to get
11
4
r
5
100
3
1/3
5 1.1
Hence
r
5 1.1 2 1 5 0.1
100
and so the IRR is 10%. The project is therefore to be recommended because this value
exceeds the market rate of 5%.
For the last part of the problem we are invited to consider whether our advice would
be different if the market rate were 12%. Using the NPV method, we need to repeat the
calculations, replacing 5 by 12. The corresponding net present value is then
$20 000(1.12)−3 − $15 000 = −$764.40
This time the NPV is negative, so the project leads to an effective loss and is not to be
recommended. The same conclusion can be reached more easily using the IRR method.
We have already seen that the internal rate of return is 10% and can deduce immedi­
ately that you would be better off investing the $15 000 at the market rate of 12%, since
this gives the higher yield.
Practice Problem
2. An investment project requires an initial outlay of $8000 and will produce a return of
$17 000 at the end of five years. Use the
(a) net present value
(b) internal rate of return
methods to decide whether this is worthwhile if the capital could be invested elsewhere
at 15% compounded annually.
Advice
This problem illustrates the use of two different methods for investment appraisal. It may appear at
first sight that the method based on the IRR is the preferred approach, particularly if you wish to
consider more than one interest rate. However, this is not usually the case. The IRR method can give
wholly misleading advice when comparing two or more projects, and you must be careful when
interpreting the results of this method. The following example highlights the difficulty.
Section 3.4 Investment appraisal 245
Example
Suppose that it is possible to invest in only one of two different projects. Project A requires
an initial outlay of $1000 and yields $1200 in four years’ time. Project B requires an outlay
of $30 000 and yields $35 000 after four years. Which of these projects would you choose
to invest in when the market rate is 3% compounded annually?
Solution
Let us first solve this problem using net present values.
For Project A
NPV = $1200(1.03)−4 − $1000 = $66.18
For Project B
NPV = $35 000(1.03)−4 − $30 000 = $1097.05
Both projects are viable as they produce positive net present values. Moreover, the second
project is preferred, since it has the higher value. You can see that this recommendation is
correct by considering how you might invest $30 000. If you opt for Project A, then the
best you can do is to invest $1000 of this amount to give a return of $1200 in four years’
time. The remaining $29 000 could be invested at the market rate of 3% to yield
$29 000(1.03)4 = $32 639.76
The total return is then
$1200 + $32 639.76 = $33 839.76
On the other hand, if you opt for Project B, then the whole of the $30 000 can be invested
to yield $35 000. In other words, in four years’ time you would be better off by choosing
Project B, which confirms the advice given by the NPV method.
However, this is contrary to the advice given by the IRR method. For Project A, the
internal rate of return, rA, satisfies
1200 5 1000 1 1
rA
100
4
Dividing by 1000 gives
11
rA
100
4
5 1.2
and if we take fourth roots, we get
11
rA
5 (1.2)1/4 5 1.047
100
so rA = 4.7%.
For Project B the internal rate of return, rB, satisfies
r
35 000 5 30 000 1 1 B
100
4
This can be solved as before to get rB = 3.9%.
Project A gives the higher internal rate of return even though, as we have seen, Project
B is the preferred choice.
246 Chapter 3 Mathematics of Finance
The results of this example show that the IRR method is an unreliable way of comparing
investment opportunities when there are significant differences between the amounts involved.
This is because the IRR method compares percentages, and obviously a large percentage of
a small sum could give a smaller profit than a small percentage of a larger sum.
Practice Problem
3. A firm needs to choose between two projects, A and B. Project A involves an initial
outlay of $13 500 and yields $18 000 in two years’ time. Project B requires an outlay of
$9000 and yields $13 000 after two years. Which of these projects would you advise the
firm to invest in if the annual market rate of interest is 7%?
So far in this section we have calculated the present value of a single future value. We now
consider the case of a sequence of payments over time. The simplest cash flow of this type
is an annuity, which is a sequence of regular equal payments. It can be thought of as the oppo­
site of a sinking fund. This time a lump sum is invested and, subsequently, equal amounts of
money are withdrawn at fixed time intervals. Provided that the payments themselves exceed
the amount of interest gained during the time interval between payments, the fund will
decrease and eventually become zero. At this point the payments cease. In practice, we are
interested in the value of the original lump sum needed to secure a regular income over a known
period of time. This can be done by summing the present values of the individual payments.
To be specific, suppose that a person wishes to retire and receive a regular income of
$10 000 at the end of each year for the next 10 years. If the interest rate is 7% compounded
annually, then the present values can be worked out using the formula
P 5 10 000 1 1
7
100
2t
The first payment is made at the end of the first year, so its present value is
P = $10 000(1.07)−1 = $9345.79
This means that if we want to take out $10 000 from the fund in one year’s time, then we
need to invest $9345.79 today. The second payment of $10 000 is made at the end of the
second year, so its present value is
$10 000(1.07)−2 = $8734.39
This is the amount of money that needs to be invested now to cover the second payment from
the fund. In general, the present value of $10 000 in t years’ time is
10 000(1.07)−t
so the total present value is
10 000(1.07)−1 + 10 000(1.07)−2 + . . . + 10 000(1.07)−10
This is a geometric series, so we may use the formula
a
rn 2 1
r 21
Section 3.4 Investment appraisal 247
In this case, a = 10 000(1.07)−1, r = 1.07−1 and n = 10, so the present value of the annuity is
$10 000(1.07)21
1.07210 2 1
5 $70 235.82
1.0721 2 1
This represents the amount of money that needs to be invested now so that a regular annual
income of $10 000 can be withdrawn from the fund for the next 10 years.
For many people there is a real fear that their pension will not provide an adequate income
for the whole of their retirement. If the income stream is to continue for ever, then we need
to investigate what happens to the formula
a
rn 2 1
r 21
as n gets bigger and bigger. In this case r = 1.07−1 < 1, so as n increases, rn decreases and tends
towards zero. This behaviour can be seen clearly from the table:
n
−n
1.07
1
10
100
0.9346
0.5083
0.0012
Setting rn = 0 in the formula for the sum of geometric series shows that if the series goes on
for ever, then eventually the sum approaches
a
12r
so that the present value of the annuity in perpetuity is
10 000(1.07)21
5 $142 857.14
1 2 1.0721
This compares with the figure of $70 235.82 calculated previously to secure the income for
10 years.
Practice Problem
4. Find the present value of an annuity that yields an income of $2000 at the end of each
month for 10 years, assuming that the interest rate is 6% compounded monthly.
The argument used in the previous example can be used to calculate the net present value.
For instance, suppose that a business requires an initial investment of $60 000, which is guar­
anteed to return a regular payment of $10 000 at the end of each year for the next 10 years.
If the discount rate is 7% compounded annually, then the previous example shows that the
present value is $70 235.82. The net present value of the investment is therefore
$70 235.82 − $60 000 = $10 235.82
A similar procedure can be used when the payments are irregular, although it is no longer
possible to use the formula for the sum of a geometric progression. Instead, the present value
of each individual payment is calculated and the values are then summed longhand.
248 Chapter 3 Mathematics of Finance
Example
A small business has a choice of investing $20 000 in one of two projects. The revenue flows
from the two projects during the next four years are listed in Table 3.26. If the interest rate
is 11% com­pounded annually, which of these two projects would you advise the company
to invest in?
Table 3.26
Revenue ($)
End of year
1
2
3
4
Total
Project A
Project B
6 000
3 000
10 000
8 000
27 000
10 000
6 000
9 000
1 000
26 000
Solution
If we simply add together all of the individual receipts, it appears that Project A is to be
preferred, since the total revenue generated from Project A is $1000 greater than that from
Project B. However, this naïve approach fails to take into account the time distribution.
From Table 3.26 we see that both projects yield a single receipt of $10 000. For Project
A this occurs at the end of Year 3, whereas for Project B this occurs at the end of Year 1.
This $10 000 is worth more in Project B because it occurs earlier in the revenue stream and,
once received, could be invested for longer at the prevailing rate of interest. To compare
these projects we need to discount the revenue stream to the present value. The present
values obtained depend on the discount rate. Table 3.27 shows the present values based on
the given rate of 11% compounded annually. These values are calculated using the formula
P = S(1.11)−t
For example, the present value of the $10 000 revenue in Project A is given by
$10 000(1.11)−3 = $7311.91
The net present values for Project A and Project B are given by
$20 422.04 − $20 000 = $422.04
and
$21 109.19 − $20 000 = $1109.19
respectively. Consequently, if it is possible to invest in only one of these projects, the
preferred choice is Project B.
Table 3.27
Discounted revenue ($)
End of year
Project A
Project B
1
2
3
4
Total
5405.41
2434.87
7311.91
5269.85
20 422.04
9000.01
4869.73
6580.72
658.73
21 109.19
Section 3.4 Investment appraisal 249
Practice Problem
5. A firm has a choice of spending $10 000 today on one of two projects. The revenue
obtained from these projects is listed in Table 3.28. Assuming that the discount rate is
15% compounded annually, which of these two projects would you advise the com­
pany to invest in?
Table 3.28
Revenue ($)
End of year
Project A
Project B
1
2
3
4
5
2000
2000
3000
3000
3000
1000
1000
2000
6000
4000
It is sometimes useful to find the internal rate of return of a project yielding a sequence of
payments over time. However, as the following example demonstrates, this can be difficult
to calculate, particularly when there are more than two payments.
Example
(a) Calculate the IRR of a project which requires an initial outlay of $20 000 and produces
a return of $8000 at the end of year 1 and $15 000 at the end of year 2.
(b) Calculate the IRR of a project which requires an initial outlay of $5000 and produces
returns of $1000, $2000 and $3000 at the end of years 1, 2 and 3, respectively.
Solution
(a) In the case of a single payment, the IRR is the annual rate of interest, r, which, when
applied to the initial outlay, P, yields a known future payment, S. If this payment is
made after t years, then
r
S 5 P 11
100
t
or, equivalently,
r
P 5 S 11
100
2t
Note that the right-hand side of this last equation is just the present value of S.
Consequently, the IRR can be thought of as the rate of interest at which the present
value of S equals the initial outlay P.
The present value of $8000 in one year’s time is
8000 1 1
r
100
21
➜
250 Chapter 3 Mathematics of Finance
where r is the annual rate of interest. Similarly, the present value of $15 000 in two
years’ time is
r
15 000 1 1
100
22
If r is to be the IRR, then the sum of these present values must equal the initial invest­
ment of $20 000. In other words, the IRR is the value of r that satisfies the equation
20 000 5 8000 1 1
r
100
21
1 15 000 1 1
r
100
22
The simplest way of solving this equation is to multiply both sides by (1 + r/100)2 to
remove all negative indices. This gives
20 000 1 1
r
100
2
5 8000 1 1
r
1 15 000
100
bm 3 bn 5 bm1n
b0 5 1
Now
11
r
100
2
5 11
r
100
11
r
r2
r
511
1
50 10 000
100
so if we multiply out the brackets, we obtain
20 000 + 400r + 2r2 = 8000 + 80r + 15 000
Collecting like terms gives
2r2 + 320r − 3000 = 0
This is a quadratic in r, so it can be solved using the formula described in Section 2.1 to get
r5
2320 6 ((320)2 2 4(2)(23000))
2(2)
2320 6 355.5
4
5 8.9% or 2168.9%
5
We can obviously ignore the negative solution, so we can conclude that the IRR is 8.9%.
(b) If an initial outlay of $5000 yields $1000, $2000 and $3000 at the end of years 1, 2 and
3, respectively, then the internal rate of return, r, satisfies the equation
5000 5 1000 1 1
r
100
21
1 2000 1 1
r
100
22
1 3000 1 1
r
100
23
A sensible thing to do here might be to multiply through by (1 + (r/100))3. However,
this produces an equation involving r3 (and lower powers of r), which is no easier to
solve than the original. Indeed, a moment’s thought should convince you that, in gen­
eral, when dealing with a sequence of payments over n years, the IRR will satisfy an
equation involving rn (and lower powers of r). Under these circumstances, it is virtually
impossible to obtain the exact solution. The best way of proceeding would be to use a
non-linear equation-solver routine on a computer, particularly if it is important that an
accurate value of r is obtained. However, if all that is needed is a rough approximation,
Section 3.4 Investment appraisal 251
then this can be done by systematic trial and error. We merely substitute likely solu­
tions into the right-hand side of the equation until we find the one that works. For
example, putting r = 5 gives
1000
200
3000
1
1
5 5358
1.05
(1.05)2 (1.05)3
Other values of the expression
1000 1 1
r
100
21
1 2000 1 1
r
100
22
1 3000 1 1
r
100
23
corresponding to r = 6, 7, . . . , 10 are listed in the following table:
r
6
7
8
9
10
value
5242
5130
5022
4917
4816
Given that we are trying to find r so that this value is 5000, this table indicates that r
is somewhere between 8% (which produces a value greater than 5000) and 9% (which
produces a value less than 5000).
If a more accurate estimate of IRR is required, then we simply try further values
between 8% and 9%. For example, it is easy to check that putting r = 8.5 gives 4969,
indicating that the exact value of r is between 8% and 8.5%. We conclude that the IRR
is 8% to the nearest percentage.
Practice Problem
6. A project requires an initial investment of $12 000. It has a guaranteed return of $8000
at the end of year 1 and a return of $2000 each year at the end of years 2, 3 and 4.
Estimate the IRR to the nearest percentage. Would you recommend that someone
invests in this project if the prevailing market rate is 8% compounded annually?
Practice Problem 6 should have convinced you how tedious it is to calculate the internal
rate of return ‘by hand’ when there are more than two payments in a revenue flow. A com­
puter spreadsheet provides the ideal tool for dealing with this. Excel’s Chart Wizard can be
used to sketch a graph from which a rough estimate of IRR can be found. A more accurate
value can be found using a ‘finer’ tabulation in the vicinity of this estimate.
We conclude this section by using the theory of discounting to explain the relationship
between interest rates and the speculative demand for money. This was first introduced in
Section 1.7 in the analysis of LM schedules. Speculative demand consists of money held in
reserve to take advantage of changes in the value of alternative financial assets, such as gov­
ernment bonds. As their name suggests, these issues can be bought from the government at a
certain price. In return, the government pays out interest on an annual basis for a prescribed
number of years. At the end of this period, the bond is redeemed and the purchaser is repaid
the original sum. Now, these bonds can be bought and sold at any point in their lifetime. The
person who chooses to buy one of these bonds part-way through this period is entitled to all
of the future interest payments, together with the final redemption payment. The value of
existing securities clearly depends on the number of years remaining before redemption,
together with the prevailing rate of interest.
252 Chapter 3 Mathematics of Finance
Example
A 10-year bond is originally offered by the government at $5000 with an annual return of
9%. Assuming that the bond has four years left before redemption, calculate its present
value assuming that the prevailing interest rate is
(a) 5% (b) 7% (c) 9% (d) 11% (e) 13%
Solution
The government pays annual interest of 9% on the $5000 bond, so agrees to pay the holder
$450 every year for 10 years. At the end of the 10 years, the bond is redeemed by the gov­
ernment and $5000 is paid back to the purchaser. If just four years remain between now
and the date of redemption, the future cash flow that is paid on the bond is summarised in
the second column of Table 3.29. This is similar to that of an annuity except that in the
final year an extra payment of $5000 is received when the government pays back the
original investment. The present value of this income stream is calculated in Table 3.29
using the given discount rates of 5%, 7%, 9%, 11% and 13% compounded annually. The
total present value in each case is given in the last row of this table and varies from $5710
when the interest rate is 5% to $4405 when it is 13%.
Notice that the value of a bond falls as interest rates rise. This is entirely to be expected,
since the formula we use to calculate individual present values is
P5
S
(1 1 r/100)t
and larger values of r produce smaller values of P.
Table 3.29
Present values
End of year
Cash flow
5%
7%
9%
11%
13%
1
2
3
4
Total present value
450
450
450
5450
429
408
389
4484
5710
421
393
367
4158
5339
413
379
347
3861
5000
405
365
329
3590
4689
398
352
312
3343
4405
The effect of this relationship on financial markets can now be analysed. Let us suppose
that the interest rate is high at, say, 13%. As you can see from Table 3.29, the price of the
bond is relatively low. Moreover, one might reasonably expect that, in the future, interest
rates are likely to fall, thereby increasing the present value of the bond. In this situation an
investor would be encouraged to buy this bond in the expectation of not only receiving the
cash flow from holding the bond but also receiving a capital gain on its present value.
Speculative balances therefore decrease as a result of high interest rates because money is
converted into securities. Exactly the opposite happens when interest rates are low. The
corresponding present value is relatively high, and, with an expectation of a rise in interest
rates and a possible capital loss, investors are reluctant to invest in securities, so specula­
tive balances are high.
Section 3.4 Investment appraisal 253
Practice Problem
7. A 10-year bond is originally offered by the government at $1000 with an annual return
of 7%. Assuming that the bond currently has three years left before redemption and that
the prevailing interest rate is 8% compounded annually, calculate its present value.
Key Terms
Annuity
time.
A lump-sum investment designed to produce a sequence of equal regular payments over
Discount rate The interest rate that is used when going backwards in time to calculate the present value from a future value.
Discounting The process of working backwards in time to find the present values from a future
value.
Internal rate of return (IRR)
Net present value (NPV)
The interest rate for which the net present value is zero.
The present value of a revenue flow minus the original cost.
Present value The amount that is invested initially to produce a specified future value after a
given period of time.
Exercise 3.4
1. Determine the present value of $7000 in two years’ time if the discount rate is 8% compounded
(a) quarterly (b) continuously
2. A small business promises a profit of $8000 on an initial investment of $20 000 after
five years.
(a) Calculate the internal rate of return.
(b) Would you advise someone to invest in this business if the market rate is 6% com­
pounded annually?
3. An investment company is considering one of two possible business ventures. Project 1
gives a return of $250 000 in four years’ time, whereas Project 2 gives a return of
$350 000 in eight years’ time. Which project should the company invest in when the
interest rate is 7% compounded annually?
4. The revenue of a firm (in $100 000s) at the end of each year for the next five years is
listed in Table 3.30. Calculate the present value of the revenue stream if the annual
discount rate is 8%.
Table 3.30
Year
1
2
3
4
5
Revenue
–20
–14
5
39
64
➜
254 Chapter 3 Mathematics of Finance
5. A builder is offered one of two methods of payment:
Option 1: A single sum of $73 000 to be paid now.
Option 2: Five equal payments of $15 000 to be paid quarterly with the first instal­
ment to be paid now.
Advise the builder which offer to accept if the interest rate is 6% compounded quarterly.
6. A financial company invests £250 000 now and receives £300 000 in three years’ time.
Calculate the internal rate of return.
7. A company has the option of investing in a project and calculates the net present values
shown in Table 3.31 at four different discount rates.
Table 3.31
Discount rate
Net present values
3
4
5
6
$5510
$630
−$3980
−$8330
(a) Estimate the internal rate of return of the project.
(b) If the money could be invested elsewhere at 5.5% interest, explain whether you
would advise the company to invest in this project.
8. You are given the opportunity of investing in one of three projects. Projects A, B and C
require initial outlays of $20 000, $30 000 and $100 000 and are guaranteed to return
$25 000, $37 000 and $117 000, respectively, in three years’ time. Which of these pro­
jects would you invest in if the market rate is 5% compounded annually?
9. Determine the present value of an annuity that pays out $100 at the end of each year
(a) for five years (b) in perpetuity
if the interest rate is 10% compounded annually.
10. An investor is given the opportunity to invest in one of two projects:
Project A costs $10 000 now and pays back $15 000 at the end of four years.
Project B costs $15 000 now and pays back $25 000 at the end of five years.
The current interest rate is 9%.
By calculating the net present values, decide which, if either, of these projects is to be
recommended.
11. A proposed investment costs $130 000 today. The expected revenue flow is $40 000 at
the end of year 1, and $140 000 at the end of year 2. Find the internal rate of return,
correct to one decimal place.
Section 3.4 Investment appraisal 255
Exercise 3.4*
1. Find the present value of $450 in six years’ time if the discount rate is 9.5% com­
pounded semi-annually. Round your answer to two decimal places.
2. A project requires an initial investment of $7000 and is guaranteed to yield a return of
$1500 at the end of the first year, $2500 at the end of the second year and $x at the end
of the third year. Find the value of x, correct to the nearest $, given that the net present
value is $838.18 when the interest rate is 6% compounded annually.
3. Determine the present value of an annuity, if it pays out $2500 at the end of each year
in perpetuity, assuming that the interest rate is 8% compounded annually.
4. A firm decides to invest in a new piece of machinery which is expected to produce an
additional revenue of $8000 at the end of every year for 10 years. At the end of this
period the firm plans to sell the machinery for scrap, for which it expects to receive
$5000. What is the maximum amount that the firm should pay for the machine if it is
not to suffer a net loss as a result of this investment? You may assume that the discount
rate is 6% compounded annually.
5. During the next three years a business decides to invest $10 000 at the beginning of each year.
The corresponding revenue that it can expect to receive at the end of each year is given in
Table 3.32. Calculate the net present value if the discount rate is 4% compounded annually.
Table 3.32
End of year
Revenue ($)
1
2
3
5 000
20 000
50 000
6. A project requires an initial investment of $50 000. It produces a return of $40 000 at the end
of year 1 and $30 000 at the end of year 2. Find the exact value of the internal rate of return.
7. A government bond that originally cost $500 with a yield of 6% has five years left
before redemption. Determine its present value if the prevailing rate of interest is 15%.
8. An annuity pays out $20 000 per year in perpetuity. If the interest rate is 5% compounded
annually, find
(a) the present value of the whole annuity;
(b) the present value of the annuity for payments received, starting from the end of the
30th year;
(c) the present value of the annuity of the first 30 years.
9. An engineering company needs to decide whether or not to build a new factory. The
costs of building the factory are $150 million initially, together with a further $100 million
at the end of the next two years. Annual operating costs are $5 million commencing at
the end of the third year. Annual revenue is predicted to be $50 million commencing at
the end of the third year. If the interest rate is 6% compounded annually, find
(a) the present value of the building costs;
(b) the present value of the operating costs at the end of n years (n > 2);
(c) the present value of the revenue after n years (n > 2);
(d) the minimum value of n for which the net present value is positive.
➜
256 Chapter 3 Mathematics of Finance
10. A project requires an initial outlay of $80 000 and produces a return of $20 000 at the
end of year 1, $30 000 at the end of year 2, and $R at the end of year 3. Determine
the value of R if the internal rate of return is 10%.
11. An annuity yields an income of $R at the end of each year for the next n years. If the inter­
est rate is r% compounded annually, show that the present value is
100R
1 2 (1 1 r/100)2n
r
(a) Find the annual income if the interest rate is 6.5%, the present value is $14 000 and
the annuity is paid for 15 years. Give your answer correct to two decimal places.
(b) Write down a general expression, in terms of r and R, for the present value if the
annuity is to be paid in perpetuity.
12. A project requiring an initial outlay of $A produces a return of $a at the end of every
year for n years.
(a) Show that the internal rate of return, r, satisfies the equation
100a
r
12 11
A5
r
100
2n
(b) Find the internal rate of return of a project which requires an initial outlay of
$1 000 000 and gives a return of $10 000 in perpetuity.
Formal mathematics
The Greek letter Σ (sigma) is the equivalent of the letter S (for sum) and is used to express
summations in a compact form. Given a sequence of numbers, x1, x2, x3, . . . , xn we define
n
S xi 5 x1 1 x2 1 x3 1 . . . 1 xn
i51
In this case we add together the terms xi starting with i = 1, allowing the subscript to go up
in integer steps, and finishing with i = n. The values of i = 1 and n which appear at the bottom
and top of the sigma sign are called the lower and upper limits, respectively. Using this notation
we have:
5
S i2 5 12 1 22 1 32 1 42 1 52
i51
A typical term is i2, so we need to add these together starting with i = 1 and running through
all integer values until we reach i = 5.
Likewise:
7
S 2i 5 23 1 24 1 25 1 26 1 27
i53
The lower limit is 3, so we begin by substituting i = 3 into the general term 2i to get 23.
Subsequent terms are obtained by substituting consecutive integer values, i = 4, 5, . . . into the
general term. The upper limit is 7, so we continue to add together all of the terms up to and
including 27.
The notation clearly provides a convenient short-hand for what would otherwise be
lengthy expressions, and there are other advantages, too. The following properties of sigma
notation allow us to manipulate summations algebraically.
Property 1
n
n
n
S (xi 1 yi) 5 S xi 1 S yi
i51
i51
i51
Property 2
n
n
S axi 5 aS xi
i51
where a is any constant
i51
These properties are easily checked by writing out the summations longhand. For Property 2
we have
n
n
i51
i51
S axi 5 ax1 1 ax2 1 . . . 1 axn 5 a(x1 1 x2 1 . . . 1 xn) 5 aS xi
As we have seen in this chapter, financial mathematics often involves the summation of series,
and the following example shows how to use it to express the sum of a geometric series using
sigma notation.
258 Chapter 3 Mathematics of Finance
Example
(a) A person saves $2000 a year in a bank account at the beginning of the year for n years.
If the bank offers a return of 4% a year, use sigma notation to write down an expres­
sion for the total amount saved.
(b) Use sigma notation to express the sum of the first n terms of a geometric series with
first term a and geometric ratio r.
Solution
(a) During a period of n years the total amount saved is
2000(1.04) + 2000(1.04)2 + 2000(1.04)3 + . . . + 2000(1.04)n
which can be expressed succinctly as
n
n
S 2000(1.04)i 5 2000S 1.04i
i51
(using property 2 to take out the factor of 2000)
i51
(b) Using sigma notation, the sum
a + ar + ar2 + . . . + arn−1
can be written as
n
n
S ar i21 5 aS r i21
i51
(again using property 2 to take out the factor of a)
i51
Key Terms
Lower limit The number which appears at the bottom of the sigma notation to indicate the first term
in a summation.
Upper limit The number which appears at the top of the sigma notation to indicate the last term in a
summation.
Multiple choice questions
Question 1
Evaluate the geometric series
200 + 200(1.05) + 200(1.05)2 + . . . . . + 200(1.05)20
Round your answer to two decimal places.
A. 7143.85
B. 6613.19
C. 7701.04
D. 6234.18
E. 6889.03
Question 2
The price of gold falls by 34% in the first six months of a year and rises by 12% in the last
six months. Work out the overall percentage decrease during the year. Round your answer
to the nearest unit.
A. 38%
B. 18%
C. 22%
D. 26%
E. 15%
Question 3
Average house prices at the start of Year 1 and Year 4 are $200 000 and $230 076. If
house prices rose by 10% in Year 1 and 5% in Year 2, work out the percentage change
in Year 3.
A. 1.2% increase
B. 1.2% decrease
C. 1.8% increase
D. 2.4% increase
E. 0.4% decrease
Question 4
Work out the missing numbers in the table below which shows the annual spending on
health care (to the nearest unit) over a four-year period and the corresponding index num­
ber (rounded to one decimal place).
Spending
460
x
512
525
Index number
89.8
95.1
100
y
A. x = 498
y = 103.8
B. x = 498
y = 102.5
C. x = 487
y = 103.8
D. x = 487
y = 102.5
E. Insufficient information
260 Chapter 3 Mathematics of Finance
Question 5
What is the future value of $5000 invested for seven years when the interest rate is 2.8%
compounded monthly?
A. $6066.27
B. $6081.25
C. $5082.24
D. $5653.33
E. $5998.34
Question 6
A regular sum of $2500 is invested at the beginning of every year at 3.5% interest com­
pounded annually. Find the total amount invested at the end of 10 years.
A. $29 328.48
B. $26 828.48
C. $30 354.98
D. $34 004.90
E. $25 921.24
Question 7
A sum of $2 million is invested today in a project which is guaranteed to provide a return
of $1.5 million at the end of five years and a further $1.25 million at the end of six years.
What is the net present value if the discount rate is 4.5% compounded annually?
A. $139 673.45
B. $167 564.09
C. $145 657.62
D. $130 000.09
E. $163 546.24
Question 8
A bank provides a mortgage of $100 000 which is to be repaid by a regular monthly repay­
ment of $x for five years. Find the value of x if the bank charges interest at 6% com­
pounded monthly.
A. $6187.57
B. $5998.78
C. $1971.20
D. $1933.28
E. $5308.65
Multiple choice questions 261
Question 9
If a principal of $3500 is invested at 3.4% interest compounded continuously, what is the
future value after six years?
A. $4277.51
B. $4917.13
C. $4292.04
D. $4290.81
E. $4276.98
Question 10
A bond is originally offered by the government for $6000 with an annual return of 6%. If
the bond has three years left before maturity, calculate its present value (correct to two
decimal places) assuming that the current interest rate is 4%.
A. $6450.98
B. $6269.90
C. $6333.01
D. $6299.07
E. $6447.79
Question 11
Find the APR of a credit card which charges interest at a monthly rate of 3.5%. Give your
answer to the nearest whole number.
A. 55%
B. 42%
C. 35%
D. 45%
E. 51%
Question 12
A piece of machinery depreciates in value by 7% a year. After how many years will the
machinery be worth half of its original price?
A. 7
B. 8
C. 9
D. 10
E. 11
262 Chapter 3 Mathematics of Finance
Question 13
A project requiring an initial outlay of $55 000 is guaranteed to produce a return of
$75 000 in three years’ time. Calculate the internal rate of return.
A. 11.1%
B. 10.7%
C. 10.9%
D. 11.4%
E. 11.6%
Question 14
Determine the present value of an annuity which pays out $12 000 at the end of each year
in perpetuity if the interest rate is 4% compounded annually.
A. $320 000
B. $312 000
C. $310 000
D. $318 000
E. $300 000
Question 15
The table shows the nominal average annual salary (in thousands of dollars) of employees
in the oil industry over a two-year period together with the rate of inflation.
Year
Salary
1
96.3
Inflation
2.8%
2
98.7
−0.5%
Work out the real average annual salary in Year 2 adjusted to prices at the beginning of
Year 1.
A. 91.2
B. 92.2
C. 93.1
D. 95.5
E. 96.5
Question 16
How much needs to be invested now to secure a regular income of $10 000 at the end of
every year for 15 years, if the interest rate is 3.25% compounded annually?
A. $118 456.87
B. $116 683.94
C. $115 432.65
D. $117 248.99
E. $115 765.89
Examination questions
Give all answers to two decimal places unless stated otherwise.
Question 1
A principal of $4000 is invested at an annual rate of 4% interest.
(a) Determine the future value after five years if the interest is compounded quarterly.
(b) Determine the future value after five years if the interest is compounded continuously.
(c) Work out the annual equivalent rate for each method of compounding in parts (a)
and (b).
(d) If the interest rate of 4% is compounded continuously, after how many years will the
investment increase to $4500?
Question 2
(a) A firm’s sales rise by 2.4% in the first six months of the year and fall by 4.9% during
the second half. Find the overall percentage change in sales over the whole year.
(b) A student invests a single amount of $P for three years in an account offering a return
of 4.8% compounded semi-annually. Find the value of P if the future value is $700.
Give your answer to the nearest cent.
(c) A person has a 25-year mortgage at a fixed rate of 6%. If the monthly repayments
required to clear the debt are $1500, work out the value of the original loan. Round
your answer to the nearest $1000.
Question 3
The number of trees in a forest at the beginning of this year was 4000. Each year 10% of
the trees are cut down and 350 new trees are planted.
(a) How many trees are in the forest at the end of (i) this year (ii) next year?
(b) Show that at the end of N years the number of trees will be 3500 + 500(0.9)N.
(c) Describe what happens to the number of trees in the long-run.
264 Chapter 3 Mathematics of Finance
Question 4
The table shows the prices of two products in two consecutive years. It also shows the
quantities bought in Year 1.
Product
Number bought in
Year 1
Unit price in
Year 1
Unit price in
Year 2
A
35
5
6
B
15
12
15
(a) Calculate the Laspeyres index using Year 1 as the base year (weighted according to the
numbers bought in Year 1).
(b) The numbers of items bought in Year 2 are shown in the following table. Use this
additional information to calculate the Paasche index.
Product
Number bought in Year 2
A
30
B
20
(c) Briefly discuss the relative merits of these two indices.
Question 5
An individual decides to invest in one of two projects, A and B, which both require the
same initial outlay of $10 000. Project A yields $11 500 in three years’ time whereas
Project B yields, $12 100 in four years’ time.
Calculate the internal rate of return for each project.
What advice can you give if the prevailing market rate of interest is 4.8%?
Question 6
(a) Country A has twice the GDP of country B. In the future, the annual growth rates of
these two countries are expected to be 1% and 4%, respectively. After how many years
will the GDP of country B exceed that of country A?
(b) The sum of $4000 is borrowed from a bank at a rate of 4.8% interest compounded
monthly. The loan is repaid in monthly instalments of $x. Show that at the end of n
months, the outstanding debt is given by 4000(1.004)n − 250x(1.004n − 1).
(i) How many months would it take to repay the loan if x = 100?
(ii) What should the monthly repayment be if the debt is to be cleared after 12 months?
(iii) Describe briefly what happens to the outstanding debt when x = 16.
Examination questions 265
Question 7
(a) A principal, P, is invested in an account with an interest rate, r%, compounded annu­
ally. The future value after n years is denoted by S.
(i)
(ii)
(iii)
(iv)
Find S when P = $4000, r = 3 and n = 6.
Find P when S = $3149.28, r = 8 and n = 3
Find r when P = $5000, n = 2 and S = $5283.92
Find n when P = $75 000, r = 10 and S = $120 788.25
(b) Find the present value of an annuity that pays out $5000 at the end of each year
(i) for 10 years
(ii) in perpetuity
if the interest rate is 4% compounded annually.
Question 8
(a) An individual saves $2000 in a bank account at the beginning of each year for 10 years.
No further savings or withdrawals are made to the account. Determine the total
amount saved if the annual interest is 3.6% compounded semi-annually.
(b) A project requiring an initial outlay of $200 000 is guaranteed to produce a return of
$145 000 in one year’s time and $80 000 in two years’ time. Calculate the net present
value of the project if the discount rate is 4.5% compounded semi-annually.
Chapter 4
Differentiation
This chapter provides a simple introduction to the general topic of calculus. There are eight
sections, which should be read in the order that they appear. It should be possible to omit
Sections 4.4 and 4.7 at a first reading, and Section 4.6 can be read any time after Section 4.3.
Section 4.1 provides a leisurely introduction to the basic idea of differentiation. The material
is explained using pictures, which will help you to understand the connection between the
underlying mathematics and the economic applications in later sections.
There are six rules of differentiation, which are evenly split between Sections 4.2 and 4.4.
Section 4.2 considers the easy rules that all students will need to know. However, if you are on
a business studies or management course, the more advanced rules in Section 4.4 may not
be of relevance and could be ignored. As far as possible, examples given in later sections and
chapters are based on the easy rules only so that such students are not disadvantaged.
However, the more advanced rules are essential to any proper study of mathematical economics,
and their use in deriving general results is unavoidable.
Sections 4.3 and 4.5 describe standard economic applications. Marginal functions associated
with revenue, cost, production, consumption and savings functions are all discussed in Section
4.3. The important topic of elasticity is described in Section 4.5. The distinction is made
between price elasticity along an arc and price elasticity at a point. Familiar results involving
general linear demand functions and the relationship between price elasticity of demand and
revenue are derived.
Sections 4.6 and 4.7 are devoted to the topic of optimisation, which is used to find the maximum
and minimum values of economic functions. In the first half of Section 4.6 we concentrate
on the mathematical technique. The second half contains four examination-type problems, all
taken from economics and business, which are solved in detail. In Section 4.7, mathematics is
used to derive general results relating to the optimisation of profit and production functions.
A simple model of stock control in business is described and a general formula for the economic
order quantity derived.
The final section revises two important mathematical functions, namely the exponential and
natural logarithm functions. We describe how to differentiate these functions and illustrate
their use in economics and business.
Differentiation is probably the most important topic in the whole text, and one that we shall
continue in Chapters 5 and 6, since it provides the necessary background theory for much of
mathematical economics. You are therefore advised to make every effort to attempt the
problems given in each section. The prerequisites include an understanding of the concept of
a function together with the ability to manipulate algebraic expressions. These are covered in
Chapters 1 and 2, and if you have worked successfully through this material, you should find
that you are in good shape to begin calculus.
Section 4.1
The derivative of a function
Objectives
At the end of this section you should be able to:
●
Find the slope of a straight line given any two points on the line.
●
Detect whether a line is uphill, downhill or horizontal using the sign of the slope.
●
Recognise the notation f′(x) and dy/dx for the derivative of a function.
●
Estimate the derivative of a function by measuring the slope of a tangent.
●
Differentiate power functions.
This introductory section is designed to get you started with differential calculus in a fairly
painless way. There are really only three things that we are going to do. We discuss the basic
idea of something called a derived function, give you two equivalent pieces of notation to
describe it, and finally show you how to write down a formula for the derived function in
simple cases.
In Chapter 1 the slope of a straight line was defined to be the change in the value of y
brought about by a 1-unit increase in x. In fact, it is not necessary to restrict the change in x
to a 1-unit increase. More generally, the slope, or gradient, of a line is taken to be the change
in y divided by the corresponding change in x as you move between any two points on the
line. It is customary to denote the change in y by Δy, where Δ is the Greek letter ‘delta’.
Likewise, the change in x is written Δx. In this notation we have
slope 5
Dy
Dx
Example
Find the slope of the straight line passing through
(a) A (1, 2) and B (3, 4) (b) A (1, 2) and C (4, 1) (c) A (1, 2) and D (5, 2)
Solution
(a) Points A and B are sketched in Figure 4.1. As we move from A to B, the y coordinate
changes from 2 to 4, which is an increase of 2 units, and the x coordinate changes from
1 to 3, which is also an increase of 2 units. Hence
slope 5
Dy
422
2
5
5 51
Dx
321
2
Section 4.1 The derivative of a function 269
y
5
B
4
Dy 5 2
3
2
A
Dx 5 2
1
0
1
2
3
4
5
x
Figure 4.1
(b) Points A and C are sketched in Figure 4.2. As we move from A to C, the y coordinate
changes from 2 to 1, which is a decrease of 1 unit, and the x coordinate changes from
1 to 4, which is an increase of 3 units. Hence
slope 5
Dy
122
21
5
5
Dx
421
3
y
3
Dx 5 3
2
A
Dy 5 21
1
C
1
2
3
4
5
x
Figure 4.2
(c) Points A and D are sketched in Figure 4.3. As we move from A to D, the y coordinate
remains fixed at 2, and the x coordinate changes from 1 to 5, which is an increase of
4 units. Hence
slope 5
Dy
222
0
5
5 50
Dx
521
4
➜
270 Chapter 4 Differentiation
y
3
Dx 5 4
Dy 5 0
2
A
D
1
1
2
3
4
5
x
Figure 4.3
Practice Problem
1. Find the slope of the straight line passing through
(a) E (–1, 3) and F (3, 11) (b) E (–1, 3) and G (4, –2) (c) E (–1, 3) and H (49, 3)
From these examples we see that the gradient is positive if the line is uphill, negative if the
line is downhill and zero if the line is horizontal.
Unfortunately, not all functions in economics are linear, so it is necessary to extend the
definition of slope to include more general curves. To do this we need the idea of a tangent,
which is illustrated in Figure 4.4.
A straight line which passes through a point on a curve and which just touches the curve
at this point is called a tangent. The slope, or gradient, of a curve at x = a is then defined to
be that of the tangent at x = a. Since we have already seen how to find the slope of a straight
line, this gives us a precise way of measuring the slope of a curve. A simple curve together
with a selection of tangents at various points is shown in Figure 4.5. Notice how each tangent
passes through exactly one point on the curve and strikes a glancing blow. In this case, the
slopes of the tangents increase as we move from left to right along the curve. This reflects the
fact that the curve is flat at x = 0 but becomes progressively steeper further away.
y
tangent at
x5a
a
Figure 4.4
y 5 f(x)
x
Section 4.1 The derivative of a function 271
y
variable slope
x
Figure 4.5
This highlights an important difference between the slope of a straight line and the slope
of a curve. In the case of a straight line, the gradient is fixed throughout its length and it is
immaterial which two points on a line are used to find it. For example, in Figure 4.6 all of the
ratios Δy/Δx have the value 1/2. However, as we have just seen, the slope of a curve varies as
we move along it. In mathematics we use the symbol
f ′(a)
read ‘f dashed of a’
to represent the slope of the graph of a function f at x = a. This notation conveys the maximum amount of information with the minimum of fuss. As usual, we need the label f to
denote which function we are considering. We certainly need the a to tell us at which point
on the curve the gradient is being measured. Finally, the ‘prime’ symbol ′ is used to distinguish the gradient from the function value. The notation f(a) gives the height of the curve
above the x axis at x = a, whereas f ′(a) gives the gradient of the curve at this point.
y
constant slope
2
1
1
2
4
1
2
x
Figure 4.6
272 Chapter 4 Differentiation
The slope of the graph of a function is called the derivative of the function. It is interesting
to notice that corresponding to each value of x there is a uniquely defined derivative f ′(x). In
other words, the rule ‘find the slope of the graph of f at x’ defines a function. This slope function
is usually referred to as the derived function. An alternative notation for the derived function is
dy
dx
read ‘dee y by dee x’
Historically, this symbol arose from the corresponding notation Δy/Δx for the gradient of a
straight line; the letter ‘d’ is the English equivalent of the Greek letter Δ. However, it is
important to realise that
dy
dx
does not mean ‘dy divided by dx’. It should be thought of as a single symbol representing the
derivative of y with respect to x. It is immaterial which notation is used, although the context
may well suggest which is more appropriate. For example, if we use
y = x2
to identify the square function, then it is natural to use
dy
dx
for the derived function. On the other hand, if we use
f(x) = x2
then f ′(x) seems more appropriate.
A graph of the square function based on the table of values
x
–2.0
f(x)
4
–1.5
2.25
–1.0
1
–0.5
0.25
0.0
0.5
1.0
1.5
2.0
0
0.25
1
2.25
4
is sketched in Figure 4.7. From this graph we see that the slopes of the tangents are
f 9(21.5) 5
21.5
5 23
0.5
f 9(20.5) 5
20.5
5 21
0.5
f ′(0) = 0
f 9(0.5) 5
0.5
51
0.5
f 9(1.5) 5
1.5
53
0.5
The value of f ′(0) is zero because the tangent is horizontal at x = 0. Notice that
f ′(−1.5) = − f ′(1.5) and f ′(−0.5) = −f ′(0.5)
Section 4.1 The derivative of a function 273
y
4
y 5 x2
3
Dy 5 1.5
Dy 5 21.5
2
Dx 5 0.5
Dx 5 0.5
1
Dy 5 0.5
Dy 5 20.5
22
21
Dx 5 0.5
Dx 5 0.5
1
2
x
horizontal
tangent at origin
Figure 4.7
This is to be expected because the graph is symmetric about the y axis. The slopes of the
tangents to the left of the y axis have the same size as those of the corresponding tangents to
the right. However, they have opposite signs since the curve slopes downhill on one side and
uphill on the other.
Practice Problem
2. Complete the following table of function values and hence sketch an accurate graph of
f (x) = x3.
x
–1.50
f (x)
x
f (x)
–1.25
–1.00
–0.75
–1.95
0.25
0.50
0.13
–0.50
–0.25
0.00
–0.13
0.75
1.00
1.25
1.50
1.95
Draw the tangents to the graph at x = −1, 0 and 1. Hence estimate the values of f ′(−1),
f ′(0) and f ′(1).
274 Chapter 4 Differentiation
Practice Problem 2 should convince you how hard it is in practice to calculate f ′(a) exactly
using graphs. It is impossible to sketch a perfectly smooth curve using graph paper and pencil,
and it is equally difficult to judge, by eye, precisely where the tangent should be. There is also
the problem of measuring the vertical and horizontal distances required for the slope of the
tangent. These inherent errors may compound to produce quite inaccurate values for f ′(a).
Fortunately, a really simple formula can be used to find f ′(a) when f is a power function. It
can be proved that
if f (x) 5 x n then f 9(x) 5 nx n21
or, equivalently,
if y 5 xn then
dy
5 nxn21
dx
The process of finding the derived function symbolically (rather than using graphs) is
known as differentiation. In order to differentiate xn all that needs to be done is to bring the
power down to the front and then to subtract 1 from the power:
subtract 1 from
the power
x n dif ferentiates to nx n−1
bring down
the power
To dif ferentiate the square function we set n = 2 in this formula to deduce that
subtract 1
f (x) = x2 dif ferentiates to f ′(x) = 2x2−1
the 2 comes down
that is,
f ′(x) = 2x1 = 2x
Using this result, we see that
f ′(−1.5) = 2 × (−1.5) = −3
f ′(−0.5) = 2 × (−0.5) = −1
f ′(0) = 2 × (0) = 0
f ′(0.5) = 2 × (0.5) = 1
f ′(1.5) = 2 × (1.5) = 3
which are in agreement with the results obtained from the graph in Figure 4.7.
Practice Problem
3. If f (x) = x3, write down a formula for f ′(x). Calculate f ′(−1), f ′(0) and f ′(1). Confirm
that these are in agreement with your rough estimates obtained in Practice Problem 2.
Section 4.1 The derivative of a function 275
Example
Differentiate
(a) y = x4 (b) y = x10 (c) y = x (d) y = 1 (e) y = 1/x4 (f) y 5 x
Solution
(a) To differentiate y = x4 we bring down the power (that is, 4) to the front and then
subtract 1 from the power (that is, 4 − 1 = 3) to deduce that
dy
5 4x3
dx
(b) Similarly,
if y 5 x10
dy
5 10x9
dx
then
(c) To use the general formula to differentiate x we first need to express y = x in the form
y = xn for some number n. In this case n = 1 because x1 = x, so
dy
5 1x0 5 1 since x0 5 1
dx
This result is also obvious from the graph of y = x sketched in Figure 4.8.
y
3
graph has a
constant slope
of 1
2
y5x
1
23
22
1
21
2
3
x
21
22
23
Figure 4.8
(d) Again, to differentiate 1 we need to express y = 1 in the form y = x n. In this case n = 0
because x 0 = 1, so
dy
5 0x21 5 0
dx
This result is also obvious from the graph of y = 1 sketched in Figure 4.9.
➜
276 Chapter 4 Differentiation
y
3
graph is horizontal
line so has a constant
slope of zero
2
y 51
1
23
22
0
21
1
2
3
x
21
22
Figure 4.9
(e) Noting that 1/x4 = x−4, it follows that
if y 5 14
then
x
dy
5 24x25 5 2 45
dx
x
The power has decreased to −5 because −4 − 1 = −5.
(f) Noting that x 5 x1/2, it follows that if
y5 x
then
dy
1
5 x21/2
dx
2
5
5
1
2x1/2
1
2 x
negative powers denote reciprocals
fractional powers denote roots
The power has decreased to 2 12 because
1
2
2 1 5 2 12 .
Practice Problem
4. Differentiate
(a) y = x5 (b) y = x6 (c) y = x100 (d) y = 1/x (e) y = 1/x2
[Hint: in parts (d) and (e) note that 1/x = x−1 and 1/x2 = x−2]
Section 4.1 The derivative of a function 277
Key Terms
Derivative The gradient of the tangent to a curve at a point. The derivative at x = a is written f ′(a).
Derived function The rule, f ′, which gives the gradient of a function, f, at a general point.
Differentiation The process or operation of determining the first derivative of a function.
Gradient The gradient of a line measures steepness and is the vertical change divided by the
horizontal change between any two points on the line. The gradient of a curve at a point is that of
the tangent at that point.
Slope
An alternative word for gradient.
Tangent
A line that just touches a curve at a point.
Exercise 4.1
1. Find the slope of the straight line passing through
(a) (2, 5) and (4, 9) (b) (3, –1) and (7, –5) (c) (7, 19) and (4, 19)
2. Verify that the points (0, 2) and (3, 0) lie on the line
2x + 3y = 6
Hence find the slope of this line. Is the line uphill, downhill or horizontal?
3. Sketch the graph of the function
f (x) = 5
Explain why it follows from this that
f ′(x) = 0
4. Differentiate the function
f (x) = x7
Hence calculate the slope of the graph of
y = x7
at the point x = 2.
5. Differentiate
(a) y = x8 (b) y = x50 (c) y = x19 (d) y = x999
6. Differentiate the following functions, giving your answer in a similar form, without
negative or fractional indices:
(a) f(x) 5
1
1
(b) f(x) 5 x (c) f(x) 5
(d) y 5 x x
3
x
x
7. Complete the following table of function values for the function, f(x) = x2 − 2x:
x
−1
−0.5
0
0.5
1
1.5
2
2.5
x − 2x
2
Sketch the graph of this function and, by measuring the slope of the tangents, estimate
(a) f ′(−0.5) (b) f ′(1) (c) f ′(1.5)
278 Chapter 4 Differentiation
Exercise 4.1*
1. Verify that the points (0, b) and (1, a + b) lie on the line
y = ax + b
Hence show that this line has slope a.
2. Differentiate each of the following functions, expressing your answer in a similar form:
1
(a) x15 (b) x4 x (c) 3 x (d)
4
x
(e)
x
x7
3. For each of the graphs
(a) y 5
x (b) y 5 x x (c) y 5
1
x
A is the point where x = 4, and B is the point where x = 4.1. In each case find
(i) the y coordinates of A and B;
(ii) the gradient of the chord AB;
(iii) the value of
dy
at A.
dx
Compare your answers to parts (ii) and (iii).
4. Find the coordinates of the point(s) at which the curve has the specified gradient.
(a) y = x2/3 gradient =
(c) y 5
1
3
1
, gradient 5 16
x2
(b) y = x5, gradient = 405
(d) y 5
1
x x
, gradient 5 2
3
64
SeCtion 4.2
Rules of differentiation
Objectives
At the end of this section you should be able to:
●
use the constant rule to differentiate a function of the form cf(x).
●
use the sum rule to differentiate a function of the form f(x) + g(x).
●
use the difference rule to differentiate a function of the form f(x) − g(x).
●
Evaluate and interpret second-order derivatives.
advice
In this section we consider three elementary rules of differentiation. Subsequent sections of this
chapter describe various applications to economics. However, before you can tackle these successfully, you must have a thorough grasp of the basic techniques involved. The problems in this section
are repetitive in nature. This is deliberate. Although the rules themselves are straightforward, it is
necessary for you to practise them over and over again before you can become proficient in using
them. In fact, mastering the rules of this section is essential for you to proceed further with the rest
of the text.
rule 1 the constant rule
If h(x) = cf (x) then h′(x) = cf ′(x)
for any constant c.
This rule tells you how to find the derivative of a constant multiple of a function:
differentiate the function and multiply by the constant
Example
Differentiate
(a) y = 2x4
(b) y = 10x
Solution
(a) To differentiate 2x4 we first differentiate x4 to get 4x3 and then multiply by 2. Hence
if y 5 2x 4 then
dy
5 2(4x 3) 5 8x 3
dx
(b) To differentiate 10x we first differentiate x to get 1 and then multiply by 10. Hence
if y 5 10x
then
dy
5 10(1) 5 10
dx
280
chaptEr 4 DIFFEREnTIATIon
practice problem
1. Differentiate
(a) y = 4x3
(b) y = 2/x
The constant rule can be used to show that
constants differentiate to zero
To see this, note that the equation
y=c
is the same as
y = cx0
because x0 = 1. By the constant rule we first differentiate x0 to get 0x−1 and then multiply by c.
Hence
if y 5 c then
dy
5 c(0x21) 5 0
dx
This result is also apparent from the graph of y = c, sketched in Figure 4.10, which is a
horizontal line c units away from the x axis. It is an important result and explains why lone
constants lurking in mathematical expressions disappear when differentiated.
rule 2 the sum rule
If h(x) = f (x) + g(x) then h′(x) = f ′(x) + g′(x)
This rule tells you how to find the derivative of the sum of two functions:
differentiate each function separately and add
y
graph is a
horizontal line so
has zero slope
y 5c
c
x
Figure 4.10
SEctiOn 4.2 RulES oF DIFFEREnTIATIon
Example
Differentiate
(a) y = x2 + x50
(b) y = x 3 + 3
Solution
(a) To differentiate x2 + x 50 we need to differentiate x2 and x 50 separately and add. Now
x2
differentiates to 2x
and
x 50
differentiates to 50x 49
so
dy
5 2x 1 50x 49
dx
if y 5 x 2 1 x 50 then
(b) To differentiate x3 + 3 we need to differentiate x3 and 3 separately and add. Now
x 3 differentiates to 3x 2
and
constants dif ferentiate
to zero
3 differentiates to 0
so
if y 5 x 3 1 3 then
dy
5 3x 2 1 0 5 3x 2
dx
practice problem
2. Differentiate
(a) y = x5 + x
(b) y = x2 + 5
rule 3 the difference rule
If h(x) = f (x) − g(x) then h′(x) = f ′(x) − g′(x)
This rule tells you how to find the derivative of the difference of two functions:
differentiate each function separately and subtract
281
282 Chapter 4 Differentiation
Example
Differentiate
(a) y = x5 − x2 (b) y 5 x 2 12
x
Solution
(a) To differentiate x5 – x2 we need to differentiate x5 and x2 separately and subtract. Now
x 5 differentiates to 5x 4
and
x 2 differentiates to 2x
so
if y 5 x 5 2 x 2 then
dy
5 5x 4 2 2x
dx
1
(b) To differentiate x 2 12 we need to differentiate x and 2 separately and subtract. Now
x
x
x differentiates to 1
and
1 differentiates to 2 2
x3
x2
x22 dif ferentiates
to 22x23
so
if y 5 x 2
1
x2
then
2
2
dy
512 2 3 511 3
dx
x
x
Practice Problem
3. Differentiate
(a) y = x2 – x3 (b) y 5 50 2
1
x3
It is possible to combine these three rules and so to find the derivative of more involved
functions, as the following example demonstrates.
Section 4.2 Rules of differentiation 283
Example
Differentiate
(a) y = 3x5 + 2x3 (b) y = x3 + 7x 2 – 2x + 10 (c) y 5 2 x 1
3
x
Solution
(a) The sum rule shows that to differentiate 3x5 + 2x3, we need to differentiate 3x5 and 2x3
separately and add. By the constant rule
3x5 differentiates to 3(5x4) = 15x4
and
2x3 differentiates to 2(3x2) = 6x2
so
if y 5 3x 5 1 2x 3 then
dy
5 15x 4 1 6x 2
dx
With practice you will soon find that you can just write the derivative down in a single
line of working by differentiating term by term. For the function
y = 3x5 + 2x3
we could just write
dy
5 3(5x 4) 1 2(3x 2) 5 15x 4 1 6x 2
dx
(b) So far we have only considered expressions comprising at most two terms. However,
the sum and difference rules still apply to lengthier expressions, so we can differentiate
term by term as before. For the function
y = x3 + 7x2 − 2x + 10
we get
dy
5 3x 2 1 7(2x ) 2 2(1) 1 0 5 3x 2 1 14x 2 2
dx
(c) To differentiate
y52 x1
3
x
we first rewrite it using the notation of indices as
y = 2x1/2 + 3x−1
Differentiating term by term then gives
dy
1 21/2
x 1 3(21)x22 5 x21/2 2 3x22
52
dx
2
which can be written in the more familiar form
1
x
2
3
x2
284 Chapter 4 Differentiation
Practice Problem
4. Differentiate
(a) y = 9x5 + 2x2
3
(b) y 5 5x 8 2 x
(c) y = x 2 + 6x + 3 (d) y = 2x4 + 12x3 – 4x2 + 7x – 400
Whenever a function is differentiated, the thing that you end up with is itself a function.
This suggests the possibility of differentiating a second time to get the ‘slope of the slope
function’. This is written as
f ″(x)
read ‘f double dashed of x’
or
d2 y
dx2
read ‘dee two y by dee x squared’
For example, if
f (x) = 5x2 − 7x + 12
then differentiating once gives
f ′(x) = 10x − 7
and if we now differentiate f ′(x) we get
f ″(x) = 10
The function f ′(x) is called the first-order derivative and f ″(x) is called the second-order derivative.
Example
Evaluate f ″(1) where
f (x) 5 x 7 1
1
x
Solution
To find f ″(1) we need to differentiate
f(x) = x7 + x−1
twice and put x = 1 into the end result. Differentiating once gives
f ′(x) = 7x6 + (−1)x−2 = 7x6 − x−2
and differentiating a second time gives
f ″(x) = 7(6x5) − (−2)x−3 = 42x5 + 2x−3
Finally, substituting x = 1 into
f 0(x) 5 42x 5 1
2
x3
gives
f ″(1) = 42 + 2 = 44
Section 4.2 Rules of differentiation 285
Practice Problem
5. Evaluate f ″(6) where
f(x) = 4x3 − 5x2
It is possible to give a graphical interpretation of the sign of the second-order derivative.
Remember that the first-order derivative, f ′(x), measures the gradient of a curve. If the deriva­
tive of f ′(x) is positive (that is, if f ″(x) > 0) then f ′(x) is increasing, so the graph gets steeper
as you move from left to right. The curve bends upwards and the function is said to be
convex. On the other hand, if f ″(x) < 0, the gradient f ′(x) must be decreasing, so the curve
bends downwards. The function is said to be concave. It is perfectly possible for a curve to
be convex for a certain range of values of x and concave for others. This is illustrated in
Figure 4.11. For this function, f ″(x) < 0 to the left of x = a, and f ″(x) > 0 to the right of x = a.
At x = a itself, the curve changes from bending downwards to bending upwards, and at this
point, f ″(a) = 0.
f 0(x) . 0
bends upwards
f 0(x) , 0
bends downwards
x 5a
Figure 4.11
The second-order derivative can be used to confirm the convexity of the general quadratic
function
f (x) = ax2 + bx + c
The first- and second-order derivatives are f ′(x) = 2ax + b and f ″(x) = 2a.
●
If a > 0, then f ″(x) > 0, so the parabola is convex.
●
If a < 0, then f ″(x) < 0, so the parabola is concave.
Of course, if a = 0, then f (x) = bx + c, which is the equation of a straight line, so the graph
bends neither upwards nor downwards.
286 Chapter 4 Differentiation
Throughout this section the functions have all been of the form y = f(x), where the letters
x and y denote the variables involved. In economic functions, different symbols are used. It
should be obvious, however, that we can still differentiate such functions by applying the
rules of this section. For example, if a supply function is given by
Q = P2 + 3P + 1
and we need to find the derivative of Q with respect to P, then we can apply the sum and
difference rules to obtain
dQ
5 2P 1 3
dP
Key Terms
Concave
Graph bends downwards when f ″(x) < 0.
Convex Graph bends upwards when f ″(x) > 0.
First-order derivative The rate of change of a function with respect to its independent variable.
It is the same as the ‘derivative’ of a function, y = f(x), and is written as f ′(x) or dy/dx.
Second-order derivative The derivative of the first-order derivative. The expression obtained
when the original function, y = f(x), is differentiated twice in succession and is written as f ″(x) or
d2y/dx2.
Exercise 4.2
1. Differentiate
3
x
(a) y = 5x2
(b) y 5
(d) y = x2 + x + 1
(e) y = x2 – 3x + 2
(f) y 5 3x 2
(g) y = 2x3 – 6x2 + 49x – 54
(h) y = ax + b
(i) y = ax2 + bx + c
( j) y 5 4x 2
(c) y = 2x + 3
7
x
3
7
1 2
x
x
2. Evaluate f ′(x) for each of the following functions at the given point:
(a) f (x) = 3x9 at x = 1
(b) f (x) = x 2 – 2x at x = 3
(c) f (x) = x3 – 4x2 + 2x – 8 at x = 0
(d) f(x) 5 5x 4 2
(e) f(x) 5 x 2
4
x4
at x 5 21
2
at x 5 4
x
Section 4.2 Rules of differentiation 287
3. By writing x 2 x 2 1 2x 2
5
5
5 x 4 1 2x 3 2 5, differentiate x 2 x 2 1 2x 2 2 .
x2
x
Use a similar approach to differentiate
(a) x2(3x – 4)
(b) x(3x3 – 2x2 + 6x – 7)
(c) (x + 1)(x – 6)
2
(d) x 2 3
x
x 2 4x 2
(e)
x3
x 2 2 3x 1 5
x2
(f)
4. Find expressions for d2y/dx2 in the case when
(a) y = 7x2 – x
(b) y 5
1
x2
(c) y = ax + b
5. Evaluate f ″(2) for the function
f (x) = x3 − 4x2 + 10x − 7
6. If f (x) = x2 – 6x + 8, evaluate f ′(3). What information does this provide about the graph
of y = f (x) at x = 3?
7. By writing 4x 5 4 3 x 5 2 x , differentiate 4x .
Use a similar approach to differentiate
(a) 25x (b) 3 27 x (c) 4 16 x 3 (d)
25
x
8. Find expressions for
(a)
dQ
for the supply function Q = P2 + P + 1
dP
(b)
d(TR)
for the total revenue function TR = 50Q – 3Q2
dQ
(c)
30
d(AC)
for the average cost function AC 5
1 10
Q
dQ
(d)
dC
for the consumption function C = 3Y + 7
dY
(e)
dQ
for the production function Q 5 10 L
dL
(f)
dp
for the profit function π = –2Q3 + 15Q2 − 24Q − 3
dQ
288 Chapter 4 Differentiation
Exercise 4.2*
1. Find the value of the first-order derivative of the function
y53 x2
81
1 13
x
when x = 9.
2. Find expressions for
dQ
for the supply function Q = 2P2 + P + 1
dP
(a)
(b) d(TR) for the total revenue function TR 5 40Q 2 3Q Q
dQ
20
(c) d(AC) for the average cost function AC 5
1 7Q 1 25
Q
dQ
(d)
dC
for the consumption function C = Y(2Y + 3) + 10
dY
(e)
dC
for the production function Q 5 200L 2 4 4 L
dL
(f)
dp
for the profit function π = –Q3 + 20Q2 − 7Q − 1
dQ
3. Find the value of the second-order derivative of the following function at the point x = 4:
f (x) = −2x3 + 4x2 + x − 3
What information does this provide about the shape of the graph of f (x) at this point?
4. Consider the graph of the function
f (x) = 2x 5 − 3x4 + 2x2 − 17x + 31
at x = −1.
Giving reasons for your answers,
(a) state whether the tangent slopes uphill, slopes downhill or is horizontal;
(b) state whether the graph is concave or convex at this point.
5. Use the second-order derivative to show that the graph of the cubic,
f (x) = ax3 + bx2 + cx + d (a > 0)
is convex when x > –b/3a and concave when x < –b/3a.
6. Find the equation of the tangent to the curve
y = 4x3 − 5x2 + x − 3
at the point where it crosses the y axis.
Section 4.2 Rules of differentiation 289
7. A Pareto income distribution function is given by
A
, x$1
xa
f(x) 5
where A and a are positive constants and x is measured in $100 000s.
(a) Find an expression for f ′(x) and hence comment on the slope of this function.
(b) Find an expression for f ″(x) and hence comment on the convexity of this function.
(c) Sketch a graph of f (x).
(d) The area under the graph between x = b and x = c measures the proportion of
people whose income is in the range, b ≤ x ≤ c. What does this graph indicate about
the distributions of income above a threshold of $100 000?
8. A utility function, U(x), measures the amount of satisfaction gained by an individual
who buys x units of a product or service. The Arrow–Pratt coefficient of relative risk
aversion is defined by
r52
xU0(x)
U9(x)
Show that the coefficient of relative risk aversion is constant for the utility function
U(x) 5
x12g
12g
SeCtion 4.3
Marginal functions
Objectives
At the end of this section you should be able to:
●
Calculate marginal revenue and marginal cost.
●
Derive the relationship between marginal and average revenue for both a monopoly and perfect
competition.
●
Calculate marginal product of labour.
●
State the law of diminishing marginal productivity using the notation of calculus.
●
Calculate marginal propensity to consume and marginal propensity to save.
At this stage you may be wondering what on earth differentiation has to do with economics.
In fact, we cannot get very far with economic theory without making use of calculus. In this
section we concentrate on three main areas that illustrate its applicability:
●
revenue and cost;
●
production;
●
consumption and savings.
We consider each of these in turn.
4.3.1 revenue and cost
In Chapter 2 we investigated the basic properties of the revenue function, TR. It is defined to
be PQ, where P denotes the price of a good and Q denotes the quantity demanded. In practice, we usually know the demand function, which provides a relationship between P and Q.
This enables a formula for TR to be written down solely in terms of Q. For example, if
P = 100 − 2Q
then
TR = PQ = (100 − 2Q)Q = 100Q − 2Q2
The formula can be used to calculate the value of TR corresponding to any value of Q. Not
content with this, we are also interested in the effect on TR of a change in the value of Q from
some existing level. To do this we introduce the concept of marginal revenue. The marginal
revenue, MR, of a good is defined by
MR 5
d(TR)
dQ
marginal revenue is the derivative of total revenue with respect to demand
Section 4.3 Marginal functions 291
For example, the marginal revenue function corresponding to
TR = 100Q − 2Q2
is given by
d(TR)
5 100 2 4Q
dQ
If the current demand is 15, say, then
MR = 100 − 4(15) = 40
You may be familiar with an alternative definition often quoted in elementary economics
textbooks. Marginal revenue is sometimes taken to be the change in TR brought about by a
1 unit increase in Q. It is easy to check that this gives an acceptable approximation to MR,
although it is not quite the same as the exact value obtained by differentiation. For example,
substituting Q = 15 into the total revenue function considered previously gives
TR = 100(15) − 2(15)2 = 1050
An increase of 1 unit in the value of Q produces a total revenue
TR = 100(16) − 2(16)2 = 1088
This is an increase of 38, which, according to the non-calculus definition, is the value of MR
when Q is 15. This compares with the exact value of 40 obtained by differentiation.
It is instructive to give a graphical interpretation of these two approaches. In Figure 4.12
the point A lies on the TR curve corresponding to a quantity Q0. The exact value of MR at
this point is equal to the derivative
d(TR)
dQ
and so is given by the slope of the tangent at A. The point B also lies on the curve but corres­
ponds to a 1-unit increase in Q. The vertical distance from A to B therefore equals the change
in TR when Q increases by 1 unit. The slope of the line joining A and B (known as a chord) is
D(TR) D(TR)
5 D(TR)
5
DQ
1
TR
B
tangent
D(TR)
A
TR curve
DQ
Q0
Figure 4.12
Q0 1 1
Q
292 Chapter 4 Differentiation
In other words, the slope of the chord is equal to the value of MR obtained from the noncalculus definition. Inspection of the diagram reveals that the slope of the tangent is approx­
imately the same as that of the chord joining A and B. In this case the slope of the tangent is
slightly the larger of the two, but there is not much in it. We therefore see that the 1-unitincrease approach produces a reasonable approximation to the exact value of MR given by
d(TR)
dQ
Practice Problem
1. If the demand function is
P = 60 − Q
find an expression for TR in terms of Q.
(1) Differentiate TR with respect to Q to find a general expression for MR in terms of
Q. Hence write down the exact value of MR at Q = 50.
(2) Calculate the value of TR when
(a) Q = 50 (b) Q = 51
and hence confirm that the 1-unit-increase approach gives a reasonable approximation to the exact value of MR obtained in part (1).
The approximation indicated by Figure 4.12 holds for any value of ΔQ. The slope of
the tangent at A is the marginal revenue, MR. The slope of the chord joining A and B is
Δ(TR)/ΔQ. It follows that
MR >
D(TR)
DQ
This equation can be transposed to give
Δ(TR) ≅ MR × ΔQ
multiply both
sides by DQ
that is,
change in total revenue > marginal revenue 3 change in demand
Moreover, Figure 4.12 shows that the smaller the value of ΔQ, the better the approximation
becomes.
Section 4.3 Marginal functions 293
Example
If the total revenue function of a good is given by
100Q − Q2
write down an expression for the marginal revenue function. If the current demand is 60,
estimate the change in the value of TR due to a 2-unit increase in Q.
Solution
If
TR = 100Q − Q2
then
MR 5
d(TR)
dQ
5 100 2 2Q
When Q = 60
MR = 100 − 2(60) = −20
If Q increases by 2 units, ΔQ = 2 and the formula
Δ(TR) ≅ MR × ΔQ
shows that the change in total revenue is approximately
(−20) × 2 = −40
A 2-unit increase in Q therefore leads to a decrease in TR of about 40.
Practice Problem
2. If the total revenue function of a good is given by
1000Q − 4Q2
write down an expression for the marginal revenue function. If the current demand is
30, find the approximate change in the value of TR due to a
(a) 3-unit increase in Q;
(b) 2-unit decrease in Q.
The simple model of demand, originally introduced in Section 1.5, assumed that price, P,
and quantity, Q, are linearly related according to an equation,
P = aQ + b
where the slope, a, is negative and the intercept, b, is positive. A downward-sloping demand
curve such as this corresponds to the case of a monopolist. A single firm, or possibly a group
294 Chapter 4 Differentiation
(a)
b
(b)
TR
AR
2b/2a
2b
2b/a
Q
MR
2b/2a
2b/a
Q
Figure 4.13
of firms forming a cartel, is assumed to be the only supplier of a particular product and
so has control over the market price. As the firm raises the price, so demand falls. The
associated total revenue function is given by
TR = PQ
= (aQ + b)Q
= aQ2 + bQ
An expression for marginal revenue is obtained by differentiating TR with respect to Q to get
MR = 2aQ + b
It is interesting to notice that, on the assumption of a linear demand equation, the marginal
revenue is also linear with the same intercept, b, but with slope 2a. The marginal revenue
curve slopes downhill exactly twice as fast as the demand curve. This is illustrated in
Figure 4.13(a).
The average revenue, AR, is defined by
AR 5
TR
Q
and, since TR = PQ, we have
AR 5
PQ
5P
Q
For this reason the demand curve is labelled average revenue in Figure 4.13(a). The above
derivation of the result AR = P is independent of the particular demand function.
Consequently, the terms ‘average revenue curve’ and ‘demand curve’ are synonymous.
Figure 4.13(a) shows that the marginal revenue takes both positive and negative values.
This is to be expected. The total revenue function is a quadratic, and its graph has the
familiar parabolic shape indicated in Figure 4.13(b). To the left of –b/2a the graph is uphill,
corresponding to a positive value of marginal revenue, whereas to the right of this point it is
downhill, giving a negative value of marginal revenue. More significantly, at the maximum
point of the TR curve, the tangent is horizontal with zero slope, and so MR is zero.
At the other extreme from a monopolist is the case of perfect competition. For this model
we assume that there are a large number of firms all selling an identical product and that
SEctiOn 4.3 MARGInAl FunCTIonS
295
AR
b
MR
Q
Figure 4.14
there are no barriers to entry into the industry. Since any individual firm produces a tiny
proportion of the total output, it has no control over price. The firm can sell only at the prevailing market price and, because the firm is relatively small, it can sell any number of goods
at this price. If the fixed price is denoted by b, then the demand function is
P=b
and the associated total revenue function is
TR = PQ = bQ
An expression for marginal revenue is obtained by differentiating TR with respect to Q and,
since b is just a constant, we see that
MR = b
In the case of perfect competition, the average and marginal revenue curves are the same.
They are horizontal straight lines, b units above the Q axis, as shown in Figure 4.14.
So far we have concentrated on the total revenue function. Exactly the same principle can
be used for other economic functions. For instance, we define the marginal cost, MC, by
MC 5
d(TC)
dQ
marginal cost is the derivative of total cost with respect to output
Again, using a simple geometrical argument, it is easy to see that if Q changes by a small
amount ΔQ, then the corresponding change in TC is given by
Δ(TC) ≅ MC × ΔQ
change in total cost > marginal cost 3 change in output
In particular, putting ΔQ = 1 gives
Δ(TC) ≅ MC
so that MC gives the approximate change in TC when Q increases by 1 unit.
296 Chapter 4 Differentiation
Example
If the average cost function of a good is
AC 5 2Q 1 6 1
13
Q
find an expression for MC. If the current output is 15, estimate the effect on TC of a 3-unit
decrease in Q.
Solution
We first need to find an expression for TC using the given formula for AC. Now we know
that the average cost is just the total cost divided by Q: that is,
AC 5
TC
Q
Hence
TC 5 (AC)Q
5 2Q 1 6 1
13
Q
Q
and, after multiplying out the brackets, we get
TC = 2Q2 + 6Q + 13
In this formula the last term, 13, is independent of Q and so must denote the fixed costs.
The remaining part, 2Q2 + 6Q, depends on Q and so represents the total variable costs.
Differentiating gives
MC 5
d(TC)
dQ
5 4Q 1 6
Notice that because the fixed costs are constant, they differentiate to zero and so have no
effect on the marginal cost. When Q = 15,
MC = 4(15) + 6 = 66
Also, if Q decreases by 2 units, then ΔQ = −2. Hence the change in TC is given by
Δ(TC) ≅ MC × ΔQ = 66 × (−2) = −132
so TC decreases by 132 units approximately.
Practice Problem
3. Find the marginal cost given the average cost function
AC 5
100
12
Q
Deduce that a 1-unit increase in Q will always result in a 2-unit increase in TC, irrespec­
tive of the current level of output.
SEctiOn 4.3 MARGInAl FunCTIonS
297
4.3.2 production
Production functions were introduced in Section 2.3. In the simplest case output, Q, is
assumed to be a function of labour, L, and capital, K. Moreover, in the short run the input
K can be assumed to be fixed, so Q is then only a function of one input L. (This is not a valid
assumption in the long run, and in general, Q must be regarded as a function of at least two
inputs. Methods for handling this situation are considered in the next chapter.) The variable
L is usually measured in terms of the number of workers or possibly in terms of the number
of worker hours. Motivated by our previous work, we define the marginal product of labour,
MPL, by
MPL 5
dQ
dL
marginal product of labour is the derivative of output with respect to labour
As before, this gives the approximate change in Q that results from using 1 more unit of L.
It is instructive to work out numerical values of MPL for the particular production function
Q = 300L1/2 − 4L
where L denotes the actual size of the workforce.
Differentiating Q with respect to L gives
dQ
dL
5 300(1/2 L21/2 ) 2 4
MPL 5
5 150L21/2 2 4
5
150
L
24
Substituting L = 1, 9, 100 and 2500 in turn into the formula for MPL gives:
(a) When L = 1
MPL 5
150
9
2 4 5 146
(b) When L = 9
MPL 5
150
1
2 4 5 46
(c) When L = 100
MPL 5
150
100
2 4 5 11
(d) When L = 2500
MPL 5
150
2500
2 4 5 21
Notice that the values of MPL decline with increasing L. Part (a) shows that if the
workforce consists of only one person then to employ two people would increase output by
298 Chapter 4 Differentiation
approximately 146. In part (b) we see that to increase the number of workers from 9 to 10
would result in about 46 additional units of output. In part (c) we see that a 1-unit increase
in labour from a level of 100 increases output by only 11. In part (d) the situation is even
worse. This indicates that to increase staff actually reduces output! The latter is a rather
surprising result, but it is borne out by what occurs in real production processes. This may
be due to problems of overcrowding on the shopfloor or to the need to create an elaborate
administration to organise the larger workforce.
This production function illustrates the law of diminishing marginal productivity (sometimes
called the law of diminishing returns). It states that the increase in output due to a 1-unit
increase in labour will eventually decline. In other words, once the size of the workforce has
reached a certain threshold level, the marginal product of labour will get smaller. For the
production function
Q = 300L1/2 − 4L
the value of MPL continually goes down with rising L. This is not always so. It is possible for
the marginal product of labour to remain constant or to go up to begin with for small values
of L. However, if it is to satisfy the law of diminishing marginal productivity, then there must
be some value of L above which MPL decreases.
A typical product curve is sketched in Figure 4.15, which has slope
dQ
5 MPL
dL
Between 0 and L0 the curve bends upwards, becoming progressively steeper, and so the slope
function, MPL, increases. Mathematically, this means that the slope of MPL is positive:
that is,
d(MPL )
.0
dQ
Now MPL is itself the derivative of Q with respect to L, so we can use the notation for the
second-order derivative and write this as
d2Q
.0
dL2
Q
bends
downwards
bends
upwards
L0
Figure 4.15
L
SEctiOn 4.3 MARGInAl FunCTIonS
299
Similarly, if L exceeds the threshold value of L0, then Figure 4.15 shows that the product
curve bends downwards and the slope decreases. In this region, the slope of the slope function is negative, so that
d2Q
,0
dL2
The law of diminishing returns states that this must happen eventually: that is,
d2Q
,0
dL2
for sufficiently large L.
practice problem
4. A Cobb–Douglas production function is given by
Q = 5L1/2K1/2
Assuming that capital, K, is fixed at 100, write down a formula for Q in terms of L
only. Calculate the marginal product of labour when
(a) L = 1
(b) L = 9
(c) L = 10 000
Verify that the law of diminishing marginal productivity holds in this case.
4.3.3 consumption and savings
In Chapter 1 the relationship between consumption, C, savings, S, and national income, Y,
was investigated. If we assume that national income is only used up in consumption and
savings then
Y=C+S
Of particular interest is the effect on C and S due to variations in Y. Expressed simply, if
national income rises by a certain amount, are people more likely to go out and spend their
extra income on consumer goods, or will they save it? To analyse this behaviour we use the
concepts marginal propensity to consume, MPC, and marginal propensity to save, MPS, which
are defined by
MPC 5
dC
dY
and
MPS 5
dS
dY
marginal propensity to consume is the derivative of consumption with respect to income
marginal propensity to save is the derivative of savings with respect to income
These definitions are consistent with those given in Section 1.7, where MPC and MPS were
taken to be the slopes of the linear consumption and savings curves, respectively. At first sight
it appears that, in general, we need to work out two derivatives in order to evaluate MPC and
300 Chapter 4 Differentiation
MPS. However, this is not strictly necessary. Recall that we can do whatever we like to an equation provided we do the same thing to both sides. Consequently, we can differentiate both
sides of the equation
Y=C+S
with respect to Y to deduce that
dY
dC
dS
5 MPC 1 MPS
5
1
dY
dY
dY
Now we are already familiar with the result that when we differentiate x with respect to x,
the answer is 1. In this case Y plays the role of x, so
dY
51
dY
Hence
1 = MPC + MPS
This formula is identical to the result given in Section 1.7 for simple linear functions. In practice, it means that we need only work out one of the derivatives. The remaining derivative
can then be calculated directly from this equation.
Example
If the consumption function is
C = 0.01Y 2 + 0.2Y + 50
calculate MPC and MPS when Y = 30.
Solution
In this example the consumption function is given, so we begin by finding MPC. To do
this we differentiate C with respect to Y. If
C = 0.01Y 2 + 0.2Y + 50
then
dC
5 0.02Y 1 0.2
dY
so, when Y = 30,
MPC = 0.02(30) + 0.2 = 0.8
To find the corresponding value of MPS we use the formula
MPC + MPS = 1
which gives
MPS = 1 − MPC = 1 − 0.8 = 0.2
This indicates that when national income increases by 1 unit (from its current level of 30),
consumption rises by approximately 0.8 units, whereas savings rise by only about 0.2 units.
At this level of income the nation has a greater propensity to consume than it has to save.
Section 4.3 Marginal functions 301
Practice Problem
5. If the savings function is given by
S = 0.02Y 2 − Y + 100
calculate the values of MPS and MPC when Y = 40. Give a brief interpretation of these results.
Key Terms
Average revenue Total revenue per unit of output: AR = TR/Q = P.
Chord A straight line joining two points on a curve.
Law of diminishing marginal productivity (law of diminishing returns) Once the size of the
workforce exceeds a particular value, the increase in output due to a 1-unit increase in labour will
decline: d2Q/dL2 < 0 for sufficiently large L.
Marginal cost The cost of producing 1 more unit of output: MC = d(TC)/dQ.
Marginal product of labour The extra output produced by 1 more unit of labour: MPL = dQ/dL.
Marginal propensity to consume The fraction of a rise in national income which goes into
consumption: MPC = dC/dY.
Marginal propensity to save The fraction of a rise in national income which goes into savings:
MPS = dS/dY.
Marginal revenue The extra revenue gained by selling 1 more unit of a good: MR = d(TR)/dQ.
Monopolist The only firm in the industry.
Perfect competition A situation in which there are no barriers to entry in an industry where there
are many firms selling an identical product at the market price.
Exercise 4.3
1. If the demand function is
P = 100 − 4Q
find expressions for TR and MR in terms of Q. Hence estimate the change in TR
brought about by a 0.3-unit increase in output from a current level of 12 units.
2. If the demand function is
P = 80 − 3Q
show that
MR = 2P − 80
3. A monopolist’s demand function is given by
P + Q = 100
Write down expressions for TR and MR in terms of Q and sketch their graphs. Find
the value of Q which gives a marginal revenue of zero and comment on the significance
of this value.
➜
302 Chapter 4 Differentiation
4. If the average cost function of a good is
AC 5
15
1 2Q 1 9
Q
find an expression for TC. What are the fixed costs in this case? Write down an expression for the marginal cost function.
5. A firm’s production function is
Q = 50L − 0.01L2
where L denotes the size of the workforce. Find the value of MPL in the case when
(a) L = 1 (b) L = 10 (c) L = 100 (d) L = 1000
Does the law of diminishing marginal productivity apply to this particular function?
6. If the consumption function is
C 5 50 1 2 Y
calculate MPC and MPS when Y = 36 and give an interpretation of these results.
7. If the consumption function is
C = 0.02Y 2 + 0.1Y + 25
find the value of Y when MPS = 0.38.
8. The price of a company’s shares, P, recorded in dollars at midday is a function of time,
t, measured in days since the beginning of the year. Give an interpretation of the statement:
dP
5 0.25
dt
when t = 6.
9. If the demand function is
P 5 3000 2 2 Q
find expressions for TR and MR. Calculate the marginal revenue when Q = 9 and give
an interpretation of this result.
Exercise 4.3*
1. A firm’s demand function is given by
P 5 100 2 4 Q 2 3Q
(a) Write down an expression for total revenue, TR, in terms of Q.
(b) Find an expression for the marginal revenue, MR, and find the value of MR when
Q = 9.
(c) Use the result of part (b) to estimate the change in TR when Q increases by 0.25
units from its current level of 9 units and compare this with the exact change in TR.
Section 4.3 Marginal functions 303
2. The consumption function is
C = 0.01Y 2 + 0.8Y + 100
(a) Calculate the values of MPC and MPS when Y = 8.
(b) Use the fact that C + S = Y to obtain a formula for S in terms of Y. By differentiating this expression, find the value of MPS at Y = 8 and verify that this agrees with
your answer to part (a).
3. The fixed costs of producing a good are 100 and the variable costs are 2 + Q/10 per unit.
(a) Find expressions for TC and MC.
(b) Evaluate MC at Q = 30 and hence estimate the change in TC brought about by a
2-unit increase in output from a current level of 30 units.
(c) At what level of output does MC = 22?
4. Show that the law of diminishing marginal productivity holds for the production function
Q = 6L2 − 0.2L3
5. A firm’s production function is given by
Q 5 5 L 2 0.1L
(a) Find an expression for the marginal product of labour, MPL.
(b) Solve the equation MPL = 0 and briefly explain the significance of this value of L.
(c) Show that the law of diminishing marginal productivity holds for this function.
6. A firm’s average cost function takes the form
AC 5 4Q 1 a 1
6
Q
and it is known that MC = 35 when Q = 3. Find the value of AC when Q = 6.
7. The total cost of producing a good is given by
TC = 250 + 20Q
The marginal revenue is 18 at Q = 219. If production is increased from its current level
of 219, would you expect profit to increase, decrease or stay the same? Give reasons for
your answer.
8. Given the demand and total cost functions
P = 150 − 2Q and TC = 40 + 0.5Q2
find the marginal profit when Q = 25 and give an interpretation of this result.
9. If the total cost function is given by TC = aQ2 + bQ + c, show that
d(AC)
MC 2 AC
5
dQ
Q
Section 4.4
Further rules of differentiation
Objectives
At the end of this section you should be able to:
●
Use the chain rule to differentiate a function of a function.
●
Use the product rule to differentiate the product of two functions.
●
Use the quotient rule to differentiate the quotient of two functions.
●
Differentiate complicated functions using a combination of rules.
Section 4.2 introduced you to the basic rules of differentiation. Unfortunately, not all functions can be differentiated using these rules alone. For example, we are unable to differentiate
the functions
x (2x 2 3)
and
x
x 11
2
using just the constant, sum or difference rules. The aim of the present section is to describe
three further rules which allow you to find the derivative of more complicated expressions.
Indeed, the totality of all six rules will enable you to differentiate any mathematical function.
Although you may find that the rules described in this section take you slightly longer to
grasp than before, they are vital to any understanding of economic theory.
The first rule that we investigate is called the chain rule, and it can be used to differentiate
functions such as
y 5 (2x 1 3)10
and y 5 (1 1 x2)
The distinguishing feature of these expressions is that they represent a ‘function of a function’. To understand what we mean by this, consider how you might evaluate
y = (2x + 3)10
on a calculator. You would first work out an intermediate number u, say, given by
u = 2x + 3
and then raise it to the power of 10 to get
y = u10
This process is illustrated using the flow chart in Figure 4.16. Note how the incoming number
x is first processed by the inner function, ‘double and add 3’. The output u from this is then
passed on to the outer function, ‘raise to the power of 10’, to produce the final outgoing
number y.
The function
y 5 (1 1 x2)
SEctiOn 4.4 FuRTHER RulES oF DIFFEREnTIATIon
x
double and
add 3
u
raise to the
power of 10
305
y
Figure 4.16
can be viewed in the same way. To calculate y you perform the inner function, ‘square and
add 1’, followed by the outer function, ‘take square roots’.
The chain rule for differentiating a function of a function may now be stated.
rule 4 the chain rule
If y is a function of u, which is itself a function of x, then
dy
dy
du
5
3
dx
du
dx
differentiate the outer function and multiply
by the derivative of the inner function
To illustrate this rule, let us return to the function
y = (2x + 3)10
in which
y = u10
and
u = 2x + 3
Now
dy
5 10u9 5 10(2x 1 3)9
du
du
52
dx
The chain rule then gives
dy
dy
du
5
3
5 10(2x 1 3)9(2) 5 20(2x 1 3)9
dx
du
dx
With practice it is possible to perform the differentiation without explicitly introducing the
variable u. To differentiate
y = (2x + 3)10
we first differentiate the outer power function to get
10(2x + 3)9
and then multiply by the derivative of the inner function, 2x + 3, which is 2, so
dy
5 20(2x 1 3)9
dx
306 Chapter 4 Differentiation
Example
Differentiate
(a) y = (3x 2 − 5x + 2)4
(b) y 5
1
3x 1 7
(c) y 5 (1 1 x 2)
Solution
(a) The chain rule shows that to differentiate (3x 2 − 5x + 2)4 we first differentiate the outer
power function to get
4(3x 2 − 5x + 2)3
and then multiply by the derivative of the inner function, 3x 2 − 5x + 2, which is 6x − 5.
Hence if
y 5 (3x2 2 5x 1 2)4
then
dy
5 4(3x 2 2 5x 1 2)3 (6x 2 5)
dx
(b) To use the chain rule to differentiate
y5
1
3x 1 7
recall that reciprocals are denoted by negative powers, so that
y = (3x + 7)−1
The outer power function differentiates to get
−(3x + 7)−2
and the inner function, 3x + 7, differentiates to get 3. By the chain rule we just multiply
these together to deduce that
if y 5
1
3x 1 7
then
dy
23
5 2(3x 1 7)22 (3) 5
dx
(3x 1 7)2
(c) To use the chain rule to differentiate
y 5 (1 1 x 2)
recall that roots are denoted by fractional powers, so that
y = (1 + x 2)1/2
The outer power function differentiates to get
1
(1 1 x 2)21/2
2
and the inner function, 1 + x 2, differentiates to get 2x. By the chain rule we just multiply these together to deduce that
if y 5 (1 1 x 2) then
1
dy
5 (1 1 x 2)21/2 (2x) 5
2
dx
x
(1 1 x 2)
SEctiOn 4.4 FuRTHER RulES oF DIFFEREnTIATIon
307
practice problem
1. Differentiate
(a) y = (3x – 4)5
(b) y = (x 2 + 3x + 5)3
(c) y 5
1
2x 2 3
(d) y 5 (4x 2 3)
The next rule is used to differentiate the product of two functions, f (x)g(x). In order to
give a clear statement of this rule, we write
u = f (x) and v = g (x)
rule 5 the product rule
dy
dv
du
5u
1v
dx
dx
dx
If y 5 uv then
This rule tells you how to differentiate the product of two functions:
multiply each function by the derivative of the other and add
Example
Differentiate
(a) y = x2(2x + 1)3
(b) x (6x 1 1)
(c) y 5
x
11x
Solution
(a) The function x 2(2x + 1)3 involves the product of two simpler functions, namely x 2 and
(2x + 1)3, which we denote by u and v, respectively. (It does not matter which function
we label u and which we label v. The same answer is obtained if u is (2x + 1)3 and v is
x 2. You might like to check this for yourself later.) Now if
u = x2
and
v = (2x + 1)3
then
du
5 2x and
dx
dv
5 6(2x 1 1)2
dx
where we have used the chain rule to find dv/dx. By the product rule,
dy
dv
du
5u
1v
dx
dx
dx
5 x2[6(2x 1 1)2] 1 (2x 1 1)3(2x)
The first term is obtained by leaving u alone and multiplying it by the derivative of v.
Similarly, the second term is obtained by leaving v alone and multiplying it by the
derivative of u.
➜
308 Chapter 4 Differentiation
If desired, the final answer may be simplified by taking out a common factor of
2x(2x + 1)2. This factor goes into the first term 3x times and into the second 2x + 1
times. Hence
dy
5 2x(2x 1 1)2[3x 1 (2x 1 1)] 5 2x(2x 1 1)2(5x 1 1)
dx
(b) The function x (6x 1 1) involves the product of the simpler functions
u 5 x and v 5 6x 1 1 5 (6x 1 1)1/2
for which
du
5 1 and
dx
dv
1
5 (6x 1 1)21/2 3 6 5 3(6x 1 1)21/2
dx
2
where we have used the chain rule to find dv/dx. By the product rule,
dy
dv
du
5u
1v
dx
dx
dx
5 x[3(6x 1 1)21/2] 1 (6x 1 1)1/2(1)
5
3x
(6x 1 1)
1 (6x 1 1)
If desired, this can be simplified by putting the second term over a common denominator
(6x 1 1)
To do this we multiply the top and bottom of the second term by 6x 1 1 to get
(6x 1 1)
(6x 1 1)
(6x 1 1) 3 (6x 1 1)
5 6x 1 1
Hence
3x 1 (6x 1 1)
dy
9x 1 1
5
5
dx
(6x 1 1)
(6x 1 1)
(c) At first sight it is hard to see how we can use the product rule to differentiate
x
11x
since it appears to be the quotient and not the product of two functions. However, if
we recall that reciprocals are equivalent to negative powers, we may rewrite it as
x(1 + x)−1
It follows that we can put
u = x and v = (1 + x)−1
which gives
du
5 1 and
dx
dv
5 2(1 1 x)22
dx
Section 4.4 Further rules of differentiation 309
where we have used the chain rule to find dv/dx. By the product rule
dy
dv
du
5u
1v
dx
dx
dx
dy
5 x[2(1 1 x)22] 1 (1 1 x)21(1)
dx
2x
1
5
1
(1 1 x)2
11x
If desired, this can be simplified by putting the second term over a common denominator
(1 + x)2
To do this we multiply the top and bottom of the second term by 1 + x to get
11x
(1 1 x)2
Hence
dy
2x
11x
2x 1 (1 1 x)
1
5
1
5
5
(1 1 x)2
(1 1 x)2
(1 1 x)2
dx (1 1 x)2
Practice Problem
2. Differentiate
(a) y = x(3x – 1)6
(b) y 5 x3 (2x 1 3)
(c) y 5
x
x22
Advice
You may have found the product rule the hardest of the rules so far. This may have been due to the
algebraic manipulation that is required to simplify the final expression. If this is the case, do not worry
about it at this stage. The important thing is that you can use the product rule to obtain some sort
of an answer even if you cannot tidy it up at the end. This is not to say that the simplification of an
expression is pointless. If the result of differentiation is to be used in a subsequent piece of theory,
it may well save time in the long run if it is simplified first.
One of the most difficult parts of Practice Problem 2 is part (c), since this involves
algebraic fractions. For this function, it is necessary to manipulate negative indices and to put
two individual fractions over a common denominator. You may feel that you are unable to
do either of these processes with confidence. For this reason we conclude this section with a
rule that is specifically designed to differentiate this type of function. The rule itself is quite
complicated. However, as will become apparent, it does the algebra for you, so you may
prefer to use it rather than the product rule when differentiating algebraic fractions.
310
chaptEr 4 DIFFEREnTIATIon
rule 6 the quotient rule
If y 5
u
v
dy
then
dx
5
vdu/dx 5 vdv/dx
v2
This rule tells you how to differentiate the quotient of two functions:
bottom times derivative of top, minus top times derivative of bottom,
all over bottom squared
Example
Differentiate
(a) y 5
x
11x
(b) y 5
1 1 x2
2 2 x3
Solution
(a) In the quotient rule, u is used as the label for the numerator and v is used for the
denominator, so to differentiate
x
11x
we must take
u=x
and
v=1+x
for which
du
5 1 and
dx
dv
51
dx
By the quotient rule
dy
vdu/dx 2 udu/dx
5
v2
dx
(1 1 x)(1) 2 x(1)
5
(1 1 x 2)
5
11x2x
(1 1 x 2)
5
1
(1 1 x 2)
Notice how the quotient rule automatically puts the final expression over a common
denominator. Compare this with the algebra required to obtain the same answer using
the product rule in part (c) of the previous example.
(b) The numerator of the algebraic fraction
1 1 x2
2 2 x3
Section 4.4 Further rules of differentiation 311
is 1 + x2 and the denominator is 2 – x3, so we take
u = 1 + x2
and
v = 2 – x3
for which
du
5 2x and
dx
dv
5 23x 2
dx
By the quotient rule
dy
vdu/dx 2 udv/dx
5
v2
dx
(2 2 x 3)(2x) 2 (1 1 x 2)(23x 2)
5
(2 2 x 3)3
5
4x 2 2x 4 1 3x 2 1 3x 4
(2 2 x3)3
5
x 4 1 3x 2 1 4x
(2 2 x 3)3
Practice Problem
3. Differentiate
(a) y 5
x 21
x
(b) y 5
x 11
x22
[You might like to check that your answer to part (a) is the same as that obtained in
Practice Problem 2(c).]
Advice
The product and quotient rules give alternative methods for the differentiation of algebraic
fractions. It does not matter which rule you go for; use whichever rule is easiest for you.
312 Chapter 4 Differentiation
Exercise 4.4
1. Use the chain rule to differentiate
(a) y = (5x + 1)3
(b) y = (2x – 7)8
(c) y = (x + 9)5
(d) y = (4x 2 – 7)3
(e) y = (x 2 + 4x – 3)4
(f) y 5 (2x 1 1)
(g) y 5
1
3x 1 1
(h) y 5
1
(4x 2 3)2
(i) y 5
1
(2x 1 5)
2. Use the product rule to differentiate
(a) y = x(3x + 4)2
(b) y = x 2(x – 2)3
(c) y 5 x (x 1 2)
(d) y = (x – 1)(x + 6)3 (e) y = (2x + 1)(x + 5)3 (f) y = x 3(2x – 5)4
3. Use the quotient rule to differentiate
(a) y 5
x
x25
(b) y 5
x
(x 1 7)
(c) y 5
x13
x22
(d) y 5
2x 1 9
3x 1 1
(e) y 5
x
(5x 1 6)
(f) y 5
x14
3x 2 7
4. Differentiate
y = (5x + 7)2
(a) by using the chain rule;
(b) by first multiplying out the brackets and then differentiating term by term.
5. Differentiate
y = x 5(x + 2)2
(a) by using the product rule;
(b) by first multiplying out the brackets and then differentiating term by term.
6. Find expressions for marginal revenue in the case when the demand function is given by
(a) P = (100 – Q)3
(b) P 5
1000
Q14
7. If the consumption function is
C5
300 1 2Y 2
11Y
calculate MPC and MPS when Y = 36 and give an interpretation of these results.
Section 4.4 Further rules of differentiation 313
Exercise 4.4*
1. Use the chain rule to differentiate
(a) y = (2x + 1)10
(d) y 5
1
x2 1 1
(b) y = (x 2 + 3x – 5)3
(c) y 5
(e) y 5 (8x 2 1)
(f) y 5
1
7x 2 3
1
3
(6x 2 5)
2. Use the product rule to differentiate
(a) y = x 2(x + 5)3
(b) y = x 5(4x + 5)2
(c) y 5 x 4(x 1 1)1/2
3. Use the quotient rule to differentiate
(a) y 5
x2
x14
(b) y 5
2x 2 1
x 11
(c) y 5
x3
(x 2 1)
4. Differentiate
(a) y = x(x – 3)4
(e) y 5
ax 1 b
cx 1 d
x
x3
(d) y 5 2
(3x 1 5)2
x 11
(b) y 5 x (2x 2 3)
(c) y 5
(f) y = (ax + b)m(cx + d)n
(g) y = x(x + 2)2(x + 3)3
5. Find an expression, simplified as far as possible, for the second-order derivative of the
function, y 5
x
.
2x 1 1
6. Find expressions for marginal revenue in the case when the demand function is given by
(a) P 5 (100 2 2Q) (b) P 5
100
21Q
7. Determine the marginal propensity to consume for the consumption function
C5
650 1 2Y 2
91Y
when Y = 21, correct to three decimal places.
Deduce the corresponding value of the marginal propensity to save and comment on
the implications of these results.
8. If the total cost function is given by
TC 5
2Q2 1 10Q
Q13
show that the marginal cost function is
MC 5 2 1
12
(Q 1 3)2
Hence comment on the behaviour of MC as Q increases.
9. If the demand function of a good is P 5 a 2 bQ 1 c show that the marginal revenue
function is
MR 5 a 2
3bQ 1 2c
2 bQ 1 c
Section 4.5
Elasticity
Objectives
At the end of this section you should be able to:
●
Calculate price elasticity averaged along an arc.
●
Calculate price elasticity evaluated at a point.
●
Decide whether supply and demand are inelastic, unit elastic or elastic.
●
Understand the relationship between price elasticity of demand and revenue.
●
Determine the price elasticity for general linear demand functions.
One important problem in business is to determine the effect on revenue of a change in the
price of a good. Let us suppose that a firm’s demand curve is downward-sloping. If the firm
lowers the price, then it will receive less for each item, but the number of items sold increases.
The formula for total revenue, TR, is
TR = PQ
and it is not immediately obvious what the net effect on TR will be as P decreases and Q
increases. The crucial factor here is not the absolute changes in P and Q but rather the proportional or percentage changes. Intuitively, we expect that if the percentage rise in Q is
greater than the percentage fall in P, then the firm experiences an increase in revenue. Under
these circumstances we say that demand is elastic, since the demand is relatively sensitive to
changes in price. Similarly, demand is said to be inelastic if demand is relatively insensitive
to price changes. In this case, the percentage change in quantity is less than the percentage
change in price. A firm can then increase revenue by raising the price of the good. Although
demand falls as a result, the increase in price more than compensates for the reduced volume
of sales and revenue rises. Of course, it could happen that the percentage changes in price and
quantity are equal, leaving revenue unchanged. We use the term unit elastic to describe this
situation.
We quantify the responsiveness of demand to price change by defining the price elasticity
of demand to be
E5
percentage change in demand
percentage change in price
Notice that because the demand curve slopes downwards, a positive change in price leads to
a negative change in quantity and vice versa. Consequently, the value of E is always negative.
It is usual for economists to ignore the negative sign and consider just the magnitude of elasticity. If this positive value is denoted by u E u, then the previous classification of demand functions can be restated more succinctly as:
SEctiOn 4.5 ElASTICITY
315
Demand is said to be
●
●
●
inelastic if u E u < 1
unit elastic if u E u = 1
elastic if u E u > 1
As usual, we denote the changes in P and Q by ΔP and ΔQ respectively, and seek a formula
for E in terms of these symbols. To motivate this, suppose that the price of a good is $12 and
that it rises to $18. A moment’s thought should convince you that the percentage change in
price is then 50%. You can probably work this out in your head without thinking too hard.
However, it is worthwhile identifying the mathematical process involved. To obtain this
figure we first express the change
18 − 12 = 6
as a fraction of the original to get
6
5 0.5
12
and then multiply by 100 to express it as a percentage. This simple example gives us a clue as
to how we might find a formula for E. In general, the percentage change in price is
change in price
expressed
essed as a fr
fraction
of the original price
DP
3 100
P
multiply by 100
to convert fractions
into percentages
Similarly, the percentage change in quantity is
DQ
3 100
Q
Hence
E5
DP
DQ
3 100
3 100 4
Q
P
Now, when we divide two fractions, we turn the denominator upside down and multiply, so
E5
5
DQ
P
3 100 3
Q
100 3 DP
DQ
P
3
Q
DP
A typical demand curve is illustrated in Figure 4.17, in which a price fall from P1 to P2 causes
an increase in demand from Q1 to Q2.
To be specific, let us suppose that the demand function is given by
P = 200 − Q2
with P1 = 136 and P2 = 119.
The corresponding values of Q1 and Q2 are obtained from the demand equation
P = 200 − Q2
by substituting P = 136 and 119, respectively, and solving for Q. For example, if P = 136, then
136 = 200 − Q2
316 Chapter 4 Differentiation
P
demand curve
P1
DP 5 P2 2 P1
P2
DQ 5 Q2 2 Q1
Q1
Q2
Q
Figure 4.17
which rearranges to give
Q2 = 200 − 136 = 64
This has solution Q = ±8 and, since we can obviously ignore the negative quantity, we have
Q1 = 8. Similarly, setting P = 119 gives Q2 = 9. The elasticity formula is
E5
P DQ
3
Q DP
and the values of ΔP and ΔQ are easily worked out to be
ΔP = 119 − 136 = −17
ΔQ = 9 − 8 = 1
However, it is not at all clear what to take for P and Q. Do we take P to be 136 or 119? Clearly
we are going to get two different answers depending on our choice. A sensible compromise
is to use their average and take
P = 1/2(136 + 119) = 127.5
Similarly, averaging the Q values gives
Q = 1/2(8 + 9) = 8.5
Hence
E5
127.5
1
5 20.88
3
8.5
217
This value is an estimate of elasticity averaged over a section of the demand curve between
(Q1, P1) and (Q2, P2). For this reason, it is called arc elasticity and is obtained by replacing P by
1
/2(P1 + P2) and Q by 1/2(Q1 + Q2) in the general formula.
Section 4.5 Elasticity 317
Practice Problem
1. Given the demand function
P = 1000 − 2Q
calculate the arc elasticity as P falls from 210 to 200.
A disappointing feature of this approach is the need to compromise and calculate the
elasticity averaged along an arc rather than calculate the exact value at a point. A formula
for the latter can easily be deduced from
E5
P DQ
3
Q DP
by considering the limit as ΔQ and ΔP tend to zero in Figure 4.17. All that happens is that
the arc shrinks to a point and the ratio ΔQ/ΔP tends to dQ/dP. The price elasticity at a point
(point elasticity) may therefore be found from
E5
P dQ
3
Q dP
Example
Given the demand function
P = 50 − 2Q
find the elasticity when the price is 30. Is demand inelastic, unit elastic or elastic at this price?
Solution
To find dQ/dP we need to differentiate Q with respect to P. However, we are actually
given a formula for P in terms of Q, so we need to transpose
P = 50 − 2Q
for Q. Adding 2Q to both sides gives
P + 2Q = 50
and if we subtract P, then
2Q = 50 − P
Finally, dividing through by 2 gives
Q = 25 − 1/2P
Hence
dQ
5 21/2
dP
➜
318 Chapter 4 Differentiation
We are given that P = 30, so at this price, demand is
Q = 25 − 1/2(30) = 10
These values can now be substituted into
E5
P dQ
3
Q dP
to get
E5
30
1
3 2 5 21.5
10
2
Moreover, since u 1.5 u > 1, demand is elastic at this price.
Practice Problem
2. Given the demand function
P = 100 − Q
calculate the magnitude of the price elasticity of demand when the price is
(a) 10 (b) 50 (c) 90
Is the demand inelastic, unit elastic or elastic at these prices?
It is quite common in economics to be given the demand function in the form
P = f(Q)
where P is a function of Q. In order to evaluate elasticity, it is necessary to find
dQ
dP
which assumes that Q is actually given as a function of P. Consequently, we may have to
transpose the demand equation and find an expression for Q in terms of P before we perform
the differentiation. This was the approach taken in the previous example. Unfortunately, if
f(Q) is a complicated expression, it may be difficult, if not impossible, to carry out the initial
rearrangement to extract Q. An alternative approach is based on the fact that
dQ
1
5
dP dP/dQ
A proof of this can be obtained via the chain rule, although we omit the details. This result
shows that we can find dQ/dP by just differentiating the original demand function to get
dP/dQ and reciprocating.
Section 4.5 Elasticity 319
Example
Given the demand function
P = −Q2 − 4Q + 96
find the price elasticity of demand when P = 51. If this price rises by 2%, calculate the
corresponding percentage change in demand.
Solution
We are given that P = 51, so to find the corresponding demand we need to solve the
quadratic equation
−Q2 − 4Q + 96 = 51
that is,
−Q2 − 4Q + 45 = 0
To do this we use the standard formula
2b 6 b2 2 4ac
2a
discussed in Section 2.1, which gives
Q5
2(24) 6 ((24)2 2 4(21)(45))
2(21)
4 6 196
22
4 6 14
5
22
5
The two solutions are –9 and 5. As usual, the negative value can be ignored, since it does
not make sense to have a negative quantity, so Q = 5.
To find the value of E we also need to calculate
dQ
dP
from the demand equation, P = −Q2 − 4Q + 96. It is not at all easy to transpose this for
Q. Indeed, we would have to use the formula for solving a quadratic, as before, replacing
the number 51 with the letter P. Unfortunately, this expression involves square roots and
the subsequent differentiation is quite messy. (You might like to have a go at this yourself!)
However, it is easy to differentiate the given expression with respect to Q to get
dP
5 22Q 2 4
dQ
and so
dQ
1
1
5
5
dP
dP/dQ 22Q 2 4
➜
320 Chapter 4 Differentiation
Finally, putting Q = 5 gives
dQ
1
52
dP
14
The price elasticity of demand is given by
E5
P dQ
3
Q dP
and if we substitute P = 51, Q = 5 and dQ/dP = –1/14, we get
E5
51
1
5 20.73
3 2
5
14
To discover the effect on Q due to a 2% rise in P, we return to the original definition
E5
percentage change in demand
percentage change in price
We know that E = –0.73 and that the percentage change in price is 2%, so
20.73 5
percentage change in demand
2%
which shows that demand changes by
−0.73 × 2% = −1.46%
A 2% rise in price therefore leads to a fall in demand of 1.46%.
Practice Problem
3. Given the demand equation
P = −Q2 − 10Q + 150
find the price elasticity of demand when Q = 4. Estimate the percentage change in price
needed to increase demand by 10%.
The price elasticity of supply is defined in an analogous way to that of demand. We define
E5
percentage change in supply
percentage change in price
An increase in price leads to an increase in supply, so E is positive.
Section 4.5 Elasticity 321
Example
Given the supply function
P 5 10 1 Q
find the price elasticity of supply
(a) averaged along an arc between Q = 100 and Q = 105;
(b) at the point Q = 100.
Solution
(a) We are given that
Q1 = 100, Q2 = 105
so that
P1 5 10 1 100 5 20 and P2 5 10 1 105 5 20.247
Hence
ΔP = 20.247 − 20 = 0.247, ΔQ = 105 − 100 = 5
P5
1
1
(20 1 20.247) 5 20.123, Q 5 (100 1 105) 5 102.5
2
2
The formula for arc elasticity gives
E5
DQ
20.123
5
P
5 3.97
3
5
3
102.5
0.247
Q
DP
(b) To evaluate the elasticity at the point Q = 100, we need to find the derivative,
supply equation
P = 10 + Q1/2
differentiates to give
dP 1 21/2
1
5
5 Q
dQ 2
2 Q
so that
dQ
52 Q
dP
At the point Q = 100, we get
dQ
5 2 100 5 20
dP
The formula for point elasticity gives
E5
dQ
20
P
3 20 5 4
3
5
dP 100
Q
Notice that, as expected, the answers to parts (a) and (b) are nearly the same.
dQ
. The
dP
322 Chapter 4 Differentiation
Practice Problem
4. If the supply equation is
Q = 150 + 5P + 0.1P2
calculate the price elasticity of supply
(a) averaged along an arc between P = 9 and P = 11;
(b) at the point P = 10.
Advice
The concept of elasticity can be applied to more general functions, and we consider some of these
in the next chapter. For the moment we investigate the theoretical properties of demand elasticity.
The following material is more difficult to understand than the foregoing, so you may prefer just to
concentrate on the conclusions and skip the intermediate derivations.
We begin by analysing the relationship between elasticity and marginal revenue. Marginal
revenue, MR, is given by
MR 5
d(TR)
dQ
Now, TR is equal to the product PQ, so we can apply the product rule to differentiate it. If
u=P
and
v=Q
then
du
dP
dv
dQ
and
51
5
5
dQ
dQ
dQ
dQ
By the product rule
MR 5 u
dv
du
1v
dQ
dQ
5P1Q3
dP
dQ
dP
Q
5P 11 3
dQ
P
check this by
multiplying out
the brackets
Now
P
dQ
5E
3
Q
dP
so
Q
dP
1
3
5
P
dQ
E
turn both sides
upside down
Section 4.5 Elasticity 323
This can be substituted into the expression for MR to get
MR 5 P 1 1
1
E
The connection between marginal revenue and demand elasticity is now complete, and this
formula can be used to justify the intuitive argument that we gave at the beginning of this
section concerning revenue and elasticity. Observe that if −1 < E < 0 then 1/E < −1, so MR
is negative for any value of P. It follows that the revenue function is decreasing in regions
where demand is inelastic, because MR determines the slope of the revenue curve. Similarly, if
E < −1, then 1/E > −1, so MR is positive for any price, P, and the revenue curve is upwards.
In other words, the revenue function is increasing in regions where demand is elastic. Finally,
if E = −1, then MR is 0, and so the slope of the revenue curve is horizontal at points where
demand is unit elastic.
Throughout this section we have taken specific functions and evaluated the elasticity
at particular points. It is more instructive to consider general functions and to deduce
general expressions for elasticity. Consider the standard linear downward-sloping demand
function
P = aQ + b
when a < 0 and b > 0. As noted in Section 4.3, this typifies the demand function faced by a
monopolist. To transpose this equation for Q, we subtract b from both sides to get
aQ = P − b
and then divide through by a to get
Q5
1
(P 2 b)
a
Hence
dQ
1
5
dP
a
The formula for elasticity of demand is
E5
P dQ
3
Q dP
so replacing Q by (1/a)(P – b) and dQ/dP by 1/a gives
E5
5
P
1
3
a
(1/a)(P 2 b)
P
P2b
Notice that this formula involves P and b but not a. Elasticity is therefore independent
of the slope of linear demand curves. In particular, this shows that, corresponding to
any price P, the elasticities of the two demand functions sketched in Figure 4.18 are
identical. This is perhaps a rather surprising result. We might have expected demand to be
more elastic at point A than at point B, since A is on the steeper curve. However, the
mathematics shows that this is not the case. (Can you explain, in economic terms, why this
is so?)
324 Chapter 4 Differentiation
P
b
elasticity is the
same at A and B
A
B
Q
Figure 4.18
Another interesting feature of the result
E5
P
P2b
is the fact that b occurs in the denominator of this fraction, so that corresponding to any
price, P, the larger the value of the intercept, b, the smaller the magnitude of the elasticity. In
Figure 4.19, the magnitude of the elasticity at C is smaller than that at D because C lies on
the curve with the larger intercept.
The dependence of E on P is also worthy of note. It shows that elasticity varies along a
linear demand curve. This is illustrated in Figure 4.20. At the left-hand end, P = b, so
E5
b
b
5 5`
0
b2b
read ‘infinity’
P
b1
magnitude of the
elasticity at C is
smaller than that at
D because b1 . b2
b2
C
D
Q
Figure 4.19
Section 4.5 Elasticity 325
P
b
|E| 5 `
|E| . 1
|E| 5 1
b/2
|E| , 1
|E| 5 0
2b/2a
2b/a
Q
Figure 4.20
At the right-hand end, P = 0, so
E5
0
0
50
5
0 2 b 2b
As you move down the demand curve, the magnitude of elasticity decreases from ∞ to 0,
taking all possible values. Demand is unit elastic when E = −1 and the price at which this
occurs can be found by solving
P
5 21 for P
P2b
P 5 b 2 P (multiply both sides by P 2 b)
2P 5 b
P5
b
2
(add P to both sides)
(divide both sides by 2)
The corresponding quantity can be found by substituting P = b/2 into the transposed demand
equation to get
Q5
1 b
b
2 b 52
a 2
2a
Demand is unit elastic exactly halfway along the demand curve. To the left of this point, u E u > 1
and demand is elastic, whereas to the right, u E u < 1 and demand is inelastic.
In our discussion of general demand functions, we have concentrated on those which are
represented by straight lines since these are commonly used in simple economic models.
There are other possibilities, and Question 4 in Exercise 4.5* investigates a class of functions
that have constant elasticity.
326 Chapter 4 Differentiation
Key Terms
Arc elasticity Elasticity measured between two points on a curve.
Elastic demand Where the percentage change in demand is more than the corresponding percentage change in price: u E u > 1.
Inelastic demand Where the percentage change in demand is less than the corresponding percentage change in price: u E u < 1.
Point elasticity Elasticity measured at a particular point on a curve: E 5
P dQ
3 .
Q dP
Price elasticity of demand A measure of the responsiveness of the change in demand due to a
change in price: (percentage change in demand) ÷ (percentage change in price).
Price elasticity of supply A measure of the responsiveness of the change in supply due to a
change in price: (percentage change in supply) ÷ (percentage change in price).
Unit elasticity of demand Where the percentage change in demand is the same as the percentage change in price: u E u = 1.
Exercise 4.5
1. Given the demand function
P = 500 − 4Q2
calculate the price elasticity of demand averaged along an arc joining Q = 8 and Q = 10.
2. Find the price elasticity of demand at the point Q = 9 for the demand function
P = 500 − 4Q2
and compare your answer with that of Question 1.
3. Find the price elasticity of demand at P = 6 for each of the following demand functions:
(a) P = 30 − 2Q
(b) P = 30 − 12Q
(c) P 5 (100 2 2Q)
4. (a) If an airline increases prices for business class flights by 8%, demand falls by about
2.5%. Estimate the elasticity of demand. Is demand elastic, inelastic or unit elastic?
(b) Explain whether you would expect a similar result to hold for economy class flights.
5. The demand function of a good is given by
Q5
1000
P2
(a) Calculate the price elasticity of demand at P = 5 and hence estimate the percentage
change in demand when P increases by 2%.
(b) Comment on the accuracy of your estimate in part (a) by calculating the exact per-
centage change in demand when P increases from 5 to 5.1.
Section 4.5 Elasticity 327
6. (a) Find the elasticity of demand in terms of Q for the demand function, P = 20 − 0.05Q.
(b) For what value of Q is demand unit elastic?
(c) Find an expression for MR and verify that MR = 0 when demand is unit elastic.
7. Consider the supply equation
Q = 4 + 0.1P2
(a) Write down an expression for dQ/dP.
(b) Show that the supply equation can be rearranged as
P 5 (10Q 2 40)
Differentiate this to find an expression for dP/dQ.
(c) Use your answers to parts (a) and (b) to verify that
dQ
1
5
dP
dP/dQ
(d) Calculate the elasticity of supply at the point Q = 14.
8. If the supply equation is
Q = 7 + 0.1P + 0.004P2
find the price elasticity of supply if the current price is 80.
(a) Is supply elastic, inelastic or unit elastic at this price?
(b) Estimate the percentage change in supply if the price rises by 5%.
Exercise 4.5*
1. Find the elasticity for the demand function
Q = 80 − 2P − 0.5P2
averaged along an arc joining Q = 32 to Q = 50. Give your answer to 2 decimal places.
2. Consider the supply equation
P = 7 + 2Q2
By evaluating the price elasticity of supply at the point P = 105, estimate the percentage
increase in supply when the price rises by 7%.
3. If the demand equation is
Q + 4P = 60
find a general expression for the price elasticity of demand in terms of P. For what value
of P is demand unit elastic?
4. Show that the price elasticity of demand is constant for the demand functions
P5
A
Qn
where A and n are positive constants.
➜
328 Chapter 4 Differentiation
5. Find a general expression for the point elasticity of supply for the function
Q = aP + b (a > 0)
Deduce that the supply function is
(a) unit elastic when b = 0;
(b) elastic when b < 0.
Give a brief geometrical interpretation of these results.
6. A supply function is given by
Q = 40 + 0.1P2
(1) Find the price elasticity of supply averaged along an arc between P = 11 and P = 13.
Give your answer correct to three decimal places.
(2) Find an expression for price elasticity of supply at a general point, P.
Hence:
(a) Estimate the percentage change in supply when the price increases by 5% from its
current level of 17. Give your answer correct to one decimal place.
(b) Find the price at which supply is unit elastic.
7. (a) Show that the elasticity of the supply function
P = aQ + b
is given by
E5
P
P2b
(b) Consider the two supply functions
P = 2Q + 5 and P = aQ + b
The quantity supplied is the same for both functions when P = 10, and at this point,
the price elasticity of supply for the second function is five times larger than that
for the first function. Find the values of a and b.
8. (a) If E denotes the elasticity of a general supply function, Q = f (P), show that the
elasticity of:
(i) Q = [ f(P)]n is nE (ii) Q = lf(P) is E (iii) Q 5 l 1 f(P) is
f(P)E
l 1 f(P)
where n and l are positive constants.
(b) Show that the elasticity of the supply function Q = P is 1 and use the results of
part (a) to write down the elasticity of
(i) Q = P3 (ii) Q 5 10P P (iii) Q 5 5 P 2 2
Section 4.6
Optimisation of economic functions
Objectives
At the end of this section you should be able to:
●
Use the first-order derivative to find the stationary points of a function.
●
Use the second-order derivative to classify the stationary points of a function.
●
Find the maximum and minimum points of an economic function.
●
Use stationary points to sketch graphs of economic functions.
In Section 2.1 a simple three-step strategy was described for sketching graphs of quadratic
functions of the form
f (x) = ax 2 + bx + c
The basic idea is to solve the corresponding equation
ax 2 + bx + c = 0
to find where the graph crosses the x axis. Provided that the quadratic equation has at least
one solution, it is then possible to deduce the coordinates of the maximum or minimum point
of the parabola. For example, if there are two solutions, then by symmetry the graph turns
round at the point exactly halfway between these solutions. Unfortunately, if the quadratic
equation has no solution, then only a limited sketch can be obtained using this approach.
In this section we show how the techniques of calculus can be used to find the coordinates
of the turning point of a parabola. The beauty of this approach is that it can be used to locate
the maximum and minimum points of any economic function, not just those represented by
quadratics. Look at the graph in Figure 4.21. Points B, C, D, E, F and G are referred to as the
stationary points (sometimes called critical points, turning points or extrema) of the function.
At a stationary point the tangent to the graph is horizontal, and so has zero slope.
Consequently, at a stationary point of a function f(x),
f ′(x) = 0
The reason for using the word ‘stationary’ is historical. Calculus was originally used by
astron­omers to predict planetary motion. If a graph of the distance travelled by an object is
sketched against time, then the speed of the object is given by the slope, since this represents
the rate of change of distance with respect to time. It follows that if the graph is horizontal
at some point, then the speed is zero and the object is instantaneously at rest: that is, stationary.
Stationary points are classified into one of three types: local maxima, local minima and
stationary points of inflection.
At a local maximum (sometimes called a relative maximum), the graph falls away on
both sides. Points B and E are the local maxima for the function sketched in Figure 4.21.
330 Chapter 4 Differentiation
y
H
B
E
D
F
C
G
A
x
Figure 4.21
The word ‘local’ is used to highlight the fact that, although these are the maximum points
relative to their locality or neighbourhood, they may not be the overall or global maximum.
In Figure 4.21 the highest point on the graph actually occurs at the right-hand end, H, which
is not a stationary point, since the slope is not zero at H.
At a local minimum (sometimes called a relative minimum), the graph rises on both sides.
Points C and G are the local minima in Figure 4.21. Again, it is not necessary for the global
minimum to be one of the local minima. In Figure 4.21 the lowest point on the graph occurs
at the left-hand end, A, which is not a stationary point.
At a stationary point of inflection, the graph rises on one side and falls on the other. The
stationary points of inflection in Figure 4.21 are labelled D and F. These points are of little
value in economics, although they do sometimes assist in sketching graphs of economic functions. Maxima and minima, on the other hand, are important. The calculation of the max­
imum points of the revenue and profit functions is clearly worthwhile. Likewise, it is useful
to be able to find the minimum points of average cost functions.
For most examples in economics, the local maximum and minimum points coincide with the
global maximum and minimum. For this reason we shall drop the word ‘local’ when describing
stationary points. However, always bear in mind that the global maximum and minimum
could actually be attained at an end point and this possibility may need to be checked. This
can be done by comparing the function values at the end points with those of the stationary
points and then deciding which of them gives rise to the largest or smallest values.
Two obvious questions remain. How do we find the stationary points of any given function, and how do we classify them? The first question is easily answered. As we mentioned
earlier, stationary points satisfy the equation
f ′(x) = 0
so all we need is to differentiate the function, to equate to zero and to solve the resulting
algebraic equation. The classification is equally straightforward. It can be shown that if a
function has a stationary point at x = a, then
●
if f ″(a) > 0, then f (x) has a minimum at x = a;
●
if f ″(a) < 0, then f (x) has a maximum at x = a.
Therefore, we need only to differentiate the function a second time and to evaluate this
second-­order derivative at each point. A point is a minimum if this value is positive and a
maximum if this value is negative. These facts are consistent with our interpretation of the
Section 4.6 Optimisation of economic functions 331
second-order derivative in Section 4.2. If f ″(a) > 0, the graph bends upwards at x = a (points
C and G in Figure 4.21). If f ″(a) < 0, the graph bends downwards at x = a (points B and E
in Figure 4.21). There is, of course, a third possibility, namely f ″(a) = 0. Sadly, when this
happens, it provides no information whatsoever about the stationary point. The point x = a
could be a maximum, minimum or inflection. This situation is illustrated in Question 2 in
Exercise 4.6* at the end of this section.
Advice
If you are unlucky enough to encounter this case, you can always classify the point by tabulating the
function values in the vicinity and use these to produce a local sketch.
To summarise, the method for finding and classifying stationary points of a function, f(x),
is as follows:
Step 1
Solve the equation f ′(x) = 0 to find the stationary points, x = a.
Step 2
If
●
f ″(a) > 0, then the function has a minimum at x = a.
●
f ″(a) < 0, then the function has a maximum at x = a.
●
f ″(a) = 0, then the point cannot be classified using the available information.
Example
Find and classify the stationary points of the following functions. Hence sketch their graphs.
(a) f (x) = x 2 − 4x + 5 (b) f(x) = 2x 3 + 3x 2 − 12x + 4
Solution
(a) In order to use steps 1 and 2 we need to find the first- and second-order derivatives of
the function
f (x) = x2 − 4x + 5
Differentiating once gives
f ′(x) = 2x − 4
and differentiating a second time gives
f ″(x) = 2
Step 1
The stationary points are the solutions of the equation
f ′(x) = 0
so we need to solve
2x − 4 = 0
➜
332 Chapter 4 Differentiation
This is a linear equation so has just one solution. Adding 4 to both sides gives
2x = 4
and dividing through by 2 shows that the stationary point occurs at
x=2
Step 2
To classify this point we need to evaluate
f ″(2)
In this case
f ″(x) = 2
for all values of x, so in particular
f ″(2) = 2
This number is positive, so the function has a minimum at x = 2.
We have shown that the minimum point occurs at x = 2. The corresponding value
of y is easily found by substituting this number into the function to get
y = (2)2 − 4(2) + 5 = 1
so the minimum point has coordinates (2, 1). A graph of f(x) is shown in Figure 4.22.
y
f(x) 5 x2 2 4x 1 5
5
(2, 1)
x
Figure 4.22
(b) In order to use steps 1 and 2 we need to find the first- and second-order derivatives of
the function
f (x) = 2x 3 + 3x 2 − 12x + 4
Differentiating once gives
f ′(x) = 6x2 + 6x − 12
and differentiating a second time gives
f ″(x) = 12x + 6
Section 4.6 Optimisation of economic functions 333
Step 1
The stationary points are the solutions of the equation
f ′(x) = 0
so we need to solve
6x 2 + 6x − 12 = 0
This is a quadratic equation and so can be solved using ‘the formula’. However, before
doing so, it is a good idea to divide both sides by 6 to avoid large numbers. The resulting equation
x2 + x − 2 = 0
has solution
x5
21 6 (12 2 4(1)(22))
21 6 9
21 6 3
5 22, 1
5
5
2
2
2(1)
In general, whenever f(x) is a cubic function, the stationary points are the solutions of
a quadratic equation, f ′(x) = 0. Moreover, we know from Section 2.1 that such an
equation can have two, one or no solutions. It follows that a cubic equation can have
two, one or no stationary points. In this particular example we have seen that there are
two stationary points, at x = −2 and x = 1.
Step 2
To classify these points we need to evaluate f ″(−2) and f ″(1). Now,
f ″(−2) = 12(−2) + 6 = −18
This is negative, so there is a maximum at x = −2. When x = −2,
y = 2(−2)3 + 3(−2)2 − 12(−2) + 4 = 24
so the maximum point has coordinates (−2, 24). Now
f ″(1) = 12(1) + 6 = 18
This is positive, so there is a minimum at x = 1. When x = 1,
y = 2(1)3 + 3(1)2 − 12(1) + 4 = −3
so the minimum point has coordinates (1, −3).
This information enables a partial sketch to be drawn as shown in Figure 4.23.
Before we can be confident about the complete picture, it is useful to plot a few more
points such as those below:
x
–10
0
10
y
–1816
4
2184
This table indicates that when x is positive, the graph falls steeply downwards from a
great height. Similarly, when x is negative, the graph quickly disappears off the bottom
of the page. The curve cannot wiggle and turn round except at the two stationary
points already plotted (otherwise it would have more stationary points, which we
know is not the case). We now have enough information to join up the pieces and so
sketch a complete picture as shown in Figure 4.24.
➜
334 Chapter 4 Differentiation
(22, 24)
y
maximum
minimum
x
(1, 23)
Figure 4.23
(22, 24)
y
f(x) 5 2x3 1 3x2 2 12x 1 4
4
x
(1, 23)
Figure 4.24
In an ideal world it would be nice to calculate the three points at which the graph
crosses the x axis. These are the solutions of
2x 3 + 3x 2 − 12x + 4 = 0
There is a formula for solving cubic equations, just as there is for quadratic equations,
but it is extremely complicated and is beyond the scope of this text.
Section 4.6 Optimisation of economic functions 335
Practice Problem
1. Find and classify the stationary points of the following functions. Hence sketch their
graphs.
(a) y = 3x 2 + 12x − 35 (b) y = −2x 3 + 15x 2 − 36x + 27
The task of finding the maximum and minimum values of a function is referred to as
optimisation. This is an important topic in mathematical economics. It provides a rich source
of examination questions, and we devote the remaining part of this section and the whole of
the next to applications of it. In this section we demonstrate the use of stationary points by
working through four ‘examination-type’ problems in detail. These problems involve the
optimisation of specific revenue, cost, profit and production functions. They are not intended
to exhaust all possibilities, although they are fairly typical. The next section describes how
the mathematics of optimisation can be used to derive general theoretical results.
Example
A firm’s short-run production function is given by
Q = 6L2 − 0.2L3
where L denotes the number of workers.
(a) Find the size of the workforce that maximises output and hence sketch a graph of this
production function.
(b) Find the size of the workforce that maximises the average product of labour. Calculate
MPL and APL at this value of L. What do you observe?
Solution
(a) In the first part of this example we want to find the value of L, which maximises
Q = 6L2 − 0.2L3
Step 1
At a stationary point
dQ
5 12L 2 0.6L2 5 0
dL
This is a quadratic equation and so we could use ‘the formula’ to find L. However, this
is not really necessary in this case because both terms have a common factor of L, and
the equation may be written as
L(12 − 0.6L) = 0
It follows that either
L = 0 or 12 − 0.6L = 0
that is, the equation has solutions
L = 0 and L = 12/0.6 = 20
➜
336 Chapter 4 Differentiation
Step 2
It is obvious on economic grounds that L = 0 is a minimum and presumably L = 20 is
the maximum. We can, of course, check this by differentiating a second time to get
d 2Q
5 12 2 1.2L
dL 2
When L = 0,
d 2Q
5 12 . 0
dL 2
which confirms that L = 0 is a minimum. The corresponding output is given by
Q = 6(0)2 − 0.2(0)3 = 0
as expected. When L = 20,
d 2Q
5 212 , 0
dL 2
which confirms that L = 20 is a maximum.
The firm should therefore employ 20 workers to achieve a maximum output
Q = 6(20)2 − 0.2(20)3 = 800
We have shown that the minimum point on the graph has coordinates (0, 0) and the
maximum point has coordinates (20, 800). There are no further turning points, so the
graph of the production function has the shape sketched in Figure 4.25.
It is possible to find the precise values of L at which the graph crosses the horizontal
axis. The production function is given by
Q = 6L2 − 0.2L3
so we need to solve
6L2 − 0.2L3 = 0
Q
(20, 800)
Q 5 6L2 2 0.2L3
(0, 0)
Figure 4.25
30
L
Section 4.6 Optimisation of economic functions 337
We can take out a factor of L2 to get
L2(6 − 0.2L) = 0
Hence, either
L2 = 0 or 6 − 0.2L = 0
The first of these merely confirms the fact that the curve passes through the origin,
whereas the second shows that the curve intersects the L axis at L = 6/0.2 = 30.
(b) In the second part of this example we want to find the value of L which maximises the
average product of labour. This is a concept that we have not met before in this text,
although it is not difficult to guess how it might be defined.
The average product of labour, APL, is taken to be total output divided by labour, so
that in symbols
APL 5
Q
L
This is sometimes called labour productivity, since it measures the average output per
worker.
In this example,
APL 5
6L 2 2 0.2L3
5 6L 2 0.2L2
L
Step 1
At a stationary point
d(APL)
50
dL
so
6 − 0.4L = 0
which has solution L = 6/0.4 = 15.
Step 2
To classify this stationary point we differentiate a second time to get
d(APL)
5 20.4 , 0
dL
which shows that it is a maximum.
The labour productivity is therefore greatest when the firm employs 15 workers. In
fact, the corresponding labour productivity, APL, is
6(15) − 0.2(15)2 = 45
In other words, the largest number of goods produced per worker is 45.
Finally, we are invited to calculate the value of MPL at this point. To find an expression for MPL we need to differentiate Q with respect to L to get
MPL = 12L − 0.6L2
When L = 15,
MPL = 12(15) − 0.6(15)2 = 45
We observe that at L = 15 the values of MPL and APL are equal.
338 Chapter 4 Differentiation
In this particular example we discovered that at the point of maximum average product of
labour
marginal product of labour 5 average product of labour
There is nothing special about this example, and in the next section we show that this result
holds for any production function.
Practice Problem
2. A firm’s short-run production function is given by
Q = 300L2 − L4
where L denotes the number of workers. Find the size of the workforce that maximises
the average product of labour and verify that at this value of L
MPL = APL
Example
The demand equation of a good is
P + Q = 30
and the total cost function is
TC = 1/2Q2 + 6Q + 7
(a) Find the level of output that maximises total revenue.
(b) Find the level of output that maximises profit. Calculate MR and MC at this value of
Q. What do you observe?
Solution
(a) In the first part of this example we want to find the value of Q which maximises total
revenue. To do this we use the given demand equation to find an expression for TR
and then apply the theory of stationary points in the usual way.
The total revenue is defined by
TR = PQ
We seek the value of Q which maximises TR, so we express TR in terms of the variable
Q only. The demand equation
P + Q = 30
can be rearranged to get
P = 30 − Q
Hence
TR = (30 − Q)Q
= 30Q − Q2
Section 4.6 Optimisation of economic functions 339
Step 1
At a stationary point
d(TR)
50
dQ
so
30 − 2Q = 0
which has the solution Q = 30/2 = 15.
Step 2
To classify this point we differentiate a second time to get
d 2(TR)
5 22
dQ 2
This is negative, so TR has a maximum at Q = 15.
(b) In the second part of this example we want to find the value of Q which maximises
profit. To do this we begin by determining an expression for profit in terms of Q. Once
this has been done, it is then a simple matter to work out the first- and second-order
derivatives and so to find and classify the stationary points of the profit function.
The profit function is defined by
π = TR − TC
From part (a)
TR = 30Q − Q2
We are given the total cost function
TC = 1/2Q2 + 6Q + 7
Hence
π = (30Q − Q2) − (1/2Q2 + 6Q + 7)
= 30Q − Q2 − 1/2Q2 − 6Q − 7
= −3/2Q2 + 24Q − 7
Step 1
At a stationary point
dp
50
dQ
so
−3Q + 24 = 0
which has solution Q = 24/3 = 8.
Step 2
To classify this point we differentiate a second time to get
d 2p
5 23
dQ2
➜
340 Chapter 4 Differentiation
This is negative, so π has a maximum at Q = 8. In fact, the corresponding maximum
profit is
π = −3/2(8)2 + 24(8) − 7 = 89
Finally, we are invited to calculate the marginal revenue and marginal cost at this
particular value of Q. To find expressions for MR and MC we need only differentiate
TR and TC, respectively. If
TR = 30Q − Q2
then
MR 5
d(TR)
dQ
5 30 2 2Q
so when Q = 8
MR = 30 − 2(8) = 14
If
TC = 1/2Q2 + 6Q + 7
then
MC 5
d(TC)
dQ
5Q16
so when Q = 8
MC = 8 + 6 = 14
We observe that at Q = 8, the values of MR and MC are equal.
In this particular example we discovered that at the point of maximum profit,
marginal revenue 5 marginal cost
There is nothing special about this example, and in the next section we show that this result
holds for any profit function.
Practice Problem
3. The demand equation of a good is given by
P + 2Q = 20
and the total cost function is
Q3 − 8Q2 + 20Q + 2
(a) Find the level of output that maximises total revenue.
(b) Find the maximum profit and the value of Q at which it is achieved. Verify that, at
this value of Q, MR = MC.
Section 4.6 Optimisation of economic functions 341
Example
The cost of building an office block, x floors high, is made up of three components:
(1) $10 million for the land;
(2) $1/4 million per floor;
(3) specialised costs of $10 000x per floor.
How many floors should the block contain if the average cost per floor is to be minimised?
Solution
The $10 million for the land is a fixed cost because it is independent of the number of
floors. Each floor costs $1/4 million, so if the building has x floors altogether, then the cost
will be 250 000x.
In addition there are specialised costs of 10 000x per floor, so if there are x floors, this will be
(10 000x)x = 10 000x 2
Notice the square term here, which means that the specialised costs rise dramatically with
increasing x. This is to be expected, since a tall building requires a more complicated
design. It may also be necessary to use more expensive materials.
The total cost, TC, is the sum of the three components: that is,
TC = 10 000 000 + 250 000x + 10 000x 2
The average cost per floor, AC, is found by dividing the total cost by the number of floors:
that is,
TC 10 000 000 1 250 000x 1 10 000x 2
5
x
x
10 000 000
1 250 000 1 10 000x
5
x
5 10 000 000x21 1 250 000 1 10 000x
AC 5
Step 1
At a stationary point
d(AC)
50
dx
In this case
210 000 000
d(AC)
5 210 000 000x22 1 10 000 5
1 10 000
dx
x2
so we need to solve
10 000 5
10 000 000
or equivalently, 10 000x 2 5 10 000 000
x2
Hence
x2 5
10 000 000
5 1000
10 000
➜
342 Chapter 4 Differentiation
This has the solution
x 5 6 1000 5 631.6
We can obviously ignore the negative value because it does not make sense to build an
office block with a negative number of floors, so we can deduce that x = 31.6.
Step 2
To confirm that this is a minimum we need to differentiate a second time. Now
d(AC)
5 210 000 000x22 1 10 000
dx
so
d 2(AC)
20 000 000
5 22(210 000 000)x23 5
2
dx
x3
When x = 31.6, we see that
d 2(AC) 20 000 000
5 633.8
5
dx 2
(31.6)3
It follows that x = 31.6 is indeed a minimum because the second-order derivative is a
positive number.
At this stage it is tempting to state that the answer is 31.6. This is mathematically
correct but is a physical impossibility since x must be a whole number. To decide whether
to take x to be 31 or 32, we simply evaluate AC for these two values of x and choose the one
that produces the lower average cost.
When x = 31,
AC 5
10 000 000
1 250 000 1 10 000(31) 5 $882 581
31
When x = 32,
AC 5
10 000 000
1 250 000 1 10 000(32) 5 $882 500
32
Therefore, an office block 32 floors high produces the lowest average cost per floor.
Practice Problem
4. The total cost function of a good is given by
TC = Q2 + 3Q + 36
Calculate the level of output that minimises average cost. Find AC and MC at this
value of Q. What do you observe?
Section 4.6 Optimisation of economic functions 343
Example
The supply and demand equations of a good are given by
P = QS + 8
and
P = −3QD + 80
respectively.
The government decides to impose a tax, t, per unit. Find the value of t which maximises the government’s total tax revenue on the assumption that equilibrium conditions
prevail in the market.
Solution
The idea of taxation was first introduced in Chapter 1. In Section 1.5 the equilibrium price
and quantity were calculated from a given value of t. In this example t is unknown, but
the analysis is exactly the same. All we need to do is to carry the letter t through the usual
calculations and then to choose t at the end so as to maximise the total tax revenue.
To take account of the tax we replace P by P − t in the supply equation. This is because
the price that the supplier actually receives is the price, P, that the consumer pays less the
tax, t, deducted by the government. The new supply equation is then
P − t = QS + 8
so that
P = QS + 8 + t
In equilibrium
QS = QD
If this common value is denoted by Q, then the supply and demand equations become
P=Q+8+t
P = −3Q + 80
Hence
Q + 8 + t = −3Q + 80
since both sides are equal to P. This can be rearranged to give
Q = −3Q + 72 − t (subtract 8 + t from both sides)
4Q = 72 − t
Q = 18 − /4t
1
(add 3Q to both sides)
(divide both sides by 4)
Now, if the number of goods sold is Q and the government raises t per good, then the total
tax revenue, T, is given by
T = tQ
= t(18 − 1/4t)
= 18t − 1/4t 2
This, then, is the expression that we wish to maximise.
➜
344 Chapter 4 Differentiation
Step 1
At a stationary point
dT
50
dt
so
18 2
1
t50
2
which has solution
t = 36
Step 2
To classify this point we differentiate a second time to get
d 2T
1
52 , 0
dt 2
2
which confirms that it is a maximum.
Hence the government should impose a tax of $36 on each good.
Practice Problem
5. The supply and demand equations of a good are given by
P = 1/2QS + 25
and
P = −2QD + 50
respectively.
The government decides to impose a tax, t, per unit. Find the value of t which
maximises the government’s total tax revenue on the assumption that equilibrium conditions prevail in the market.
In theory a spreadsheet such as Excel could be used to solve optimisation problems,
although it cannot handle the associated mathematics. The preferred method is to use a
symbolic computation system such as Maple, Matlab, Mathcad or Derive which can not only
sketch the graphs of functions but also differentiate and solve equations. Consequently, it is
possible to obtain the exact solution using one of these packages.
Section 4.6 Optimisation of economic functions 345
Key Terms
Average product of labour (labour productivity)
Output per worker: APL = Q/L.
Maximum (local) point A point on a curve which has the highest function value in comparison
with other values in its neighbourhood; at such a point the first-order derivative is zero and the
second-order derivative is either zero or negative.
Minimum (local) point A point on a curve which has the lowest function value in comparison with
other values in its neighbourhood; at such a point the first-order derivative is zero and the secondorder derivative is either zero or positive.
Optimisation The determination of the optimal (usually stationary) points of a function.
Stationary point of inflection A stationary point that is neither a maximum nor a minimum; at
such a point both the first- and second-order derivatives are zero.
Stationary points (critical points, turning points, extrema) Points on a graph at which the
tangent is horizontal; at a stationary point the first-order derivative is zero.
Exercise 4.6
1. Find and classify the stationary points of the following functions. Hence give a rough
sketch of their graphs.
(a) y = −x 2 + x + 1 (b) y = x 2 − 4x + 4 (c) y = x 2 − 20x + 105 (d) y = −x 3 + 3x
2. If the demand equation of a good is
P = 40 − 2Q
find the level of output that maximises total revenue.
3. A firm’s short-run production function is given by
Q = 30L2 − 0.5L3
Find the value of L which maximises APL and verify that MPL = APL at this point.
4. If the fixed costs are 13 and the variable costs are Q + 2 per unit, show that the average
cost function is
AC 5
13
1Q12
Q
(a) Calculate the values of AC when Q = 1, 2, 3, . . . , 6. Plot these points on graph
paper and hence produce an accurate graph of AC against Q.
(b) Use your graph to estimate the minimum average cost.
(c) Use differentiation to confirm your estimate obtained in part (b).
5. The demand and total cost functions of a good are
4P + Q − 16 = 0
and
TC 5 4 1 2Q 2
respectively.
3Q2
Q3
1
10
20
➜
346 Chapter 4 Differentiation
(a) Find expressions for TR, π, MR and MC in terms of Q.
(b) Solve the equation
dp
50
dQ
and hence determine the value of Q which maximises profit.
(c) Verify that, at the point of maximum profit, MR = MC.
6. The supply and demand equations of a good are given by
3P − QS = 3
and
2P + QD = 14
respectively.
The government decides to impose a tax, t, per unit. Find the value of t which
maximises the government’s total tax revenue on the assumption that equilibrium conditions prevail in the market.
7. A manufacturer has fixed costs of $200 each week, and the variable costs per unit can
be expressed by the function, VC = 2Q − 36.
(a) Find an expression for the total cost function and deduce that the average cost
function is given by
AC 5
200
1 2Q 2 36
Q
(b) Find the stationary point of this function and show that this is a minimum.
(c) Verify that, at this stationary point, average cost is the same as marginal cost.
8. A firm’s short-run production function is given by
Q53 L
where L is the number of units of labour.
If the price per unit sold is $50 and the price per unit of labour is $10, find the value
of L needed to maximise profits. You may assume that the firm sells all that it produces,
and you can ignore all other costs.
9. The average cost per person of hiring a tour guide on a week’s river cruise for a
maximum party size of 30 people is given by
AC = 3Q2 − 192Q + 3500 (0 < Q ≤ 30)
Find the minimum average cost for the trip.
10. An electronic components firm launches a new product on 1 January. During the
following year a rough estimate of the number of orders, S, received t days after the
launch is given by
S = t2 − 0.002t3
What is the maximum number of orders received on any one day of the year?
Section 4.6 Optimisation of economic functions 347
Exercise 4.6*
1. A firm’s demand function is
P = 60 − 0.5Q
If fixed costs are 10 and variable costs are Q + 3 per unit, find the maximum profit.
2. Show that all of the following functions have a stationary point at x = 0. Verify in each
case that f ″(0) = 0. Classify these points by producing a rough sketch of each function.
(a) f (x) = x3 (b) f (x) = x 4 (c) f(x) = −x 6
3. If fixed costs are 15 and the variable costs are 2Q per unit, write down expressions for
TC, AC and MC. Find the value of Q which minimises AC and verify that AC = MC at
this point.
4. Daily sales, S, of a new product for the first two weeks after the launch is modelled by
S = t3 − 24t2 + 180t + 60 (0 ≤ t ≤ 13)
where t is the number of days.
(a) Find and classify the stationary points of this function.
(b) Sketch a graph of S against t on the interval 0 ≤ t ≤ 13.
(c) Find the maximum and minimum daily sales during the period between t = 5 and
t = 9.
5. If the demand function of a good is
P 5 (1000 2 4Q)
find the value of Q which maximises total revenue.
6. A firm’s total cost and demand functions are given by
TC = Q2 + 50Q + 10 and P = 200 − 4Q
respectively.
(a) Find the level of output needed to maximise the firm’s profit.
(b) The government imposes a tax of $t per good. If the firm adds this tax to its costs
and continues to maximise profit, show that the price of the good increases by twofifths of the tax, irrespective of the value of t.
7. Given that the cubic function f(x) = x 3 + ax 2 + bx + c has a stationary point at (2, 5)
and that it passes through (1, 3), find the values of a, b and c.
8. The total revenue function of a good is given by:
TR = 0.2Q3 on 0 ≤ Q ≤ 5
TR = −4Q2 + 55Q − 150 on 5 ≤ Q ≤ 10
(a) Sketch a graph of TR against Q on the interval 0 ≤ Q ≤ 10.
(b) Find the maximum revenue and the value of Q at which it is achieved.
(c) For what value of Q is the marginal revenue a maximum?
Section 4.7
Further optimisation of economic
functions
Objectives
At the end of this section you should be able to:
●
Show that, at the point of maximum profit, marginal revenue equals marginal cost.
●
Show that, at the point of maximum profit, the slope of the marginal revenue curve is less than
that of marginal cost.
●
Maximise profits of a firm with and without price discrimination in different markets.
●
Show that, at the point of maximum average product of labour, average product of labour equals
marginal product of labour.
●
Derive a formula for the economic order quantity in stock control.
The previous section demonstrated how mathematics can be used to optimise particular
economic functions. Those examples suggested two important results:
1. If a firm maximises profit, then MR = MC.
2. If a firm maximises average product of labour, then APL = MPL.
Although these results were found to hold for all of the examples considered in Section 4.6,
it does not necessarily follow that the results are always true. The aim of this section is to
prove these assertions without reference to specific functions and hence to demonstrate their
generality.
Advice
You may prefer to skip these proofs at a first reading and just concentrate on the worked example
(as well as Practice Problem 2 and Question 3 in Exercise 4.7*) on price discrimination.
Justification of the first result turns out to be really quite easy. Profit, π, is defined to be
the difference between total revenue, TR, and total cost, TC: that is,
π = TR − TC
To find the stationary points of π we differentiate with respect to Q and equate to zero: that is,
d(TR) d(TC)
dp
50
5
2
dQ
dQ
dQ
Section 4.7 Further optimisation of economic functions 349
where we have used the difference rule to differentiate the right-hand side. In Section 4.3 we defined
MR 5
d(TR)
d(TC)
and MC 5
dQ
dQ
so the previous equation is equivalent to
MR − MC = 0
and so MR = MC as required.
The stationary points of the profit function can therefore be found by sketching the MR
and MC curves on the same diagram and inspecting the points of intersection. Figure 4.26
shows typical marginal revenue and marginal cost curves. The result
MR = MC
holds for any stationary point. Consequently, if this equation has more than one solution,
then we need some further information before we can decide on the profit-maximising level
of output. In Figure 4.26 there are two points of intersection, Q1 and Q2, and it turns out (as
you discovered in Practice Problem 3 and Question 5 of Exercise 4.6 in the previous section)
that one of these is a maximum while the other is a minimum. Obviously, in any actual
example, we can classify these points by evaluating second-order derivatives. However, it
would be nice to make this decision just by inspecting the graphs of marginal revenue and
marginal cost. To see how this can be done, let us return to the equation
dp
5 MR 2 MC
dQ
and differentiate again with respect to Q to get
d2p
d(MR) d(MC)
5
2
2
dQ
dQ
dQ
Now, if d2π/dQ2 < 0, then the profit is a maximum. This will be so when
d(MR)
d(MC)
,
dQ
dQ
MC
MR
Q1
Figure 4.26
Q2
Q
350 Chapter 4 Differentiation
that is, when the slope of the marginal revenue curve is less than the slope of the marginal
cost curve.
Looking at Figure 4.26, we deduce that this criterion is true at Q2, so this must be the
desired level of output needed to maximise profit. Note also from Figure 4.26 that the statement
‘the slope of the marginal revenue curve is less than the slope of the marginal cost curve’ is equival­
ent to saying that ‘the marginal cost curve cuts the marginal revenue curve from below’. It is
this latter form that is often quoted in economics textbooks. A similar argument shows that,
at a minimum point, the marginal cost curve cuts the marginal revenue curve from above, and
so we can deduce that profit is a minimum at Q1 in Figure 4.26. In practice, the task of sketching
the graphs of MR and MC and reading off the coordinates of the points of intersection is not
an attractive one, particularly if MR and MC are complicated functions. However, it might
turn out that MR and MC are both linear, in which case a graphical approach is feasible.
Practice Problem
1. A monopolist’s demand function is
P = 25 − 0.5Q
The fixed costs of production are 7 and the variable costs are Q + 1 per unit.
(a) Show that
TR = 25Q − 0.5Q2 and TC = Q2 + Q + 7
and deduce the corresponding expressions for MR and MC.
(b) Sketch the graphs of MR and MC on the same diagram and hence find the value of
Q which maximises profit.
Quite often a firm identifies more than one market in which it wishes to sell its goods. For
example, a firm might decide to export goods to several countries and demand conditions are
likely to be different in each one. The firm may be able to take advantage of this and increase
overall profit by charging different prices in each country. The theoretical result ‘marginal revenue
equals marginal cost’ can be applied in each market separately to find the optimal pricing policy.
Example
A firm is allowed to charge different prices for its domestic and industrial customers. If P1
and Q1 denote the price and demand for the domestic market, then the demand equation is
P1 + Q1 = 500
If P2 and Q2 denote the price and demand for the industrial market, then the demand
equation is
2P2 + 3Q2 = 720
The total cost function is
TC = 50 000 + 20Q
where Q = Q1 + Q2. Determine the prices (in dollars) that the firm should charge to maximise profits:
Section 4.7 Further optimisation of economic functions 351
(a) with price discrimination;
(b) without price discrimination.
Compare the profits obtained in parts (a) and (b).
Solution
(a) The important thing to notice is that the total cost function is independent of the
market and so marginal costs are the same in each case. In fact, since
TC = 50 000 + 20Q
we have MC = 20. All we have to do to maximise profits is to find an expression for
the marginal revenue for each market and to equate this to the constant value of marginal cost.
Domestic market
The demand equation
P1 + Q1 = 500
rearranges to give
P1 = 500 − Q1
so the total revenue function for this market is
TR1 = (500 − Q1)Q1 = 500Q1 − Q21
Hence
MR1 5
d(TR1)
5 500 2 2Q1
dQ1
For maximum profit
MR1 = MC
so
500 − 2Q1 = 20
which has solution Q1 = 240. The corresponding price is found by substituting this
value into the demand equation to get
P1 = 500 − 240 = $260
To maximise profit the firm should charge its domestic customers $260 per good.
Industrial market
The demand equation
2P2 + 3Q2 = 720
rearranges to give
P2 = 360 − 3/2Q2
so the total revenue function for this market is
TR2 = (360 − 3/2Q2)Q2 = 360Q − 3/2Q22
➜
352 Chapter 4 Differentiation
Hence
MR2 5
d(TR2)
5 360 2 3Q2
dQ2
For maximum profit
MR2 = MC
so
360 − 3Q2 = 20
which has solution Q2 = 340/3. The corresponding price is obtained by substituting this
value into the demand equation to get
P2 5 360 2
3 340
5 $190
2 3
To maximise profits the firm should charge its industrial customers $190 per good,
which is lower than the price charged to its domestic customers.
(b) If there is no price discrimination, then P1 = P2 = P, say, and the demand functions for
the domestic and industrial markets become
P + Q1 = 500
and
2P + 3Q2 = 720
respectively. We can use these to deduce a single demand equation for the combined
market. We need to relate the price, P, of each good to the total demand, Q = Q1 + Q2.
This can be done by rearranging the given demand equations for Q1 and Q2 and then
adding. For the domestic market
Q1 = 500 − P
and for the industrial market
Q2 = 240 − 2/3P
Hence
Q = Q1 + Q2 = 740 − 5/3P
The demand equation for the combined market is therefore
Q + 5/3P = 740
The usual procedure for profit maximisation can now be applied. This demand equation rearranges to give
P = 444 − 3/5Q
enabling the total revenue function to be written as
3
3Q 2
TR 5 444 2 Q Q 5 444Q 2
5
5
Section 4.7 Further optimisation of economic functions 353
Hence
MR 5
d(TR)
6
5 444 2 Q
dQ
5
For maximum profit
MR = MC
so
444 2
6
Q 5 20
5
which has solution Q = 1060/3. The corresponding price is found by substituting this
value into the demand equation to get
P 5 444 2
3 1060
5 $232
5
3
To maximise profit without discrimination, the firm needs to charge a uniform price
of $232 for each good. Notice that this price lies between the prices charged to its
domestic and industrial customers with discrimination.
To evaluate the profit under each policy we need to work out the total revenue and
subtract the total cost. In part (a) the firm sells 240 goods at $260 each in the domestic
market and sells 340/3 goods at $190 each in the industrial market, so the total revenue
received is
240 3 260 1
340
3 190 5 $83 933.33
3
The total number of goods produced is
240 1
340 1060
5
3
3
so the total cost is
50 000 1 20 3
1060
5 $57 066.67
3
Therefore, the profit with price discrimination is
83 933.33 − 57 066.67 = $26 866.67
In part (b) the firm sells 1060/3 goods at $232 each, so total revenue is
1060
3 232 5 $81 973.33
3
Now the total number of goods produced under both pricing policies is the same: that
is, 1060/3. Consequently, the total cost of production in part (b) must be the same as
part (a): that is,
TC = $57 066.67
The profit without price discrimination is
81 973.33 − 57 066.67 = $24 906.66
As expected, the profits are higher with discrimination than without.
354 Chapter 4 Differentiation
Practice Problem
2. A firm has the possibility of charging different prices in its domestic and foreign
markets. The corresponding demand equations are given by
Q1 = 300 − P1
Q2 = 400 − 2P2
The total cost function is
TC = 5000 + 100Q
where Q = Q1 + Q2.
Determine the prices (in dollars) that the firm should charge to maximise profits
(a) with price discrimination;
(b) without price discrimination.
Compare the profits obtained in parts (a) and (b).
In the previous example and in Practice Problem 2 we assumed that the marginal costs
were the same in each market. The level of output that maximises profit with price discrim­
ination was found by equating marginal revenue to this common value of marginal cost. It
follows that the marginal revenue must be the same in each market. In symbols,
MR1 = MC and MR2 = MC
so
MR1 = MR2
This fact is obvious on economic grounds. If it were not true, then the firm’s policy would be to
increase sales in the market where marginal revenue is higher and to decrease sales by the same
amount in the market where the marginal revenue is lower. The effect would be to increase
revenue while keeping costs fixed, thereby raising profit. This property leads to an interesting
result connecting price, P, with elasticity of demand, E. In Section 4.5 we derived the formula
MR 5 P 1 1
1
E
If we let the price elasticity of demand in two markets be denoted by E1 and E2 corresponding
to prices P1 and P2, then the equation
MR1 = MR2
becomes
P1 1 1
1
1
5 P2 1 1
E1
E2
This equation holds whenever a firm chooses its prices P1 and P2 to maximise profits in each
market. Note that if u E1 u < u E2 u, then this equation can be true only if P1 > P2. In other
words, the firm charges the higher price in the market where the magnitude of the elasticity
of demand is lower.
Section 4.7 Further optimisation of economic functions 355
Practice Problem
3. Calculate the price elasticity of demand at the point of maximum profit for each of the
demand functions given in Practice Problem 2 with price discrimination. Verify that
the firm charges the higher price in the market with the lower value of u E u.
The previous discussion concentrated on profit. We now turn our attention to average product of labour and prove result (2) stated at the beginning of this section. This concept is
defined by
APL 5
Q
L
where Q is output and L is labour. The maximisation of APL is a little more complicated
than before, since it is necessary to use the quotient rule to differentiate this function. In the
notation of Section 4.4 we write
u = Q and v = L
so
du
dQ
5 MPL
5
dL
dL
and
dv
dL
51
5
dL
dL
where we have used the fact that the derivative of output with respect to labour is the
marginal product of labour.
The quotient rule gives
d(APL)
vdu / dL 2 udv / dL
5
dL
v2
L(MPL) 2 Q(1)
5
L2
5
MPL 2 Q / L
L
5
MPL 2 APL
L
divide top and
bottom by L
by definition,
Q
APL 5
L
At a stationary point
d(APL )
50
dL
Hence
MPL = APL
as required.
This analysis shows that, at a stationary point of the average product of labour function, the
marginal product of labour equals the average product of labour. The above argument provides
a formal proof that this result is true for any average product of labour function. Figure 4.27
356 Chapter 4 Differentiation
MPL
APL
L0
L
Figure 4.27
shows typical average and marginal product functions. Note that the two curves intersect at
the peak of the APL curve. To the left of this point the APL function is increasing, so that
d(APL)
.0
dL
Now, we have just seen that
MPL 2 APL
d(APL)
5
dL
L
so we deduce that, to the left of the maximum, MPL > APL. In other words, in this region the
graph of marginal product of labour lies above that of average product of labour. Similarly,
to the right of the maximum, APL is decreasing, so that
d(APL)
,0
dL
and hence MPL < APL. The graph of marginal product of labour therefore lies below that of
average product of labour in this region.
We deduce that if the stationary point is a maximum, then the MPL curve cuts the APL
curve from above. A similar argument can be used for any average function. The particular
case of the average cost function is investigated in Question 8 in Exercise 4.7*.
We conclude this section by investigating a simple model of inventory control in business.
Most firms need to buy and store goods during the year. A furniture manufacturer has to
obtain wood and other raw materials to make its goods, and a grocery store needs to have a
regular supply of frozen food to store in its freezers. Management needs to make decisions
about when to order more stock and the size of each order. Let us suppose that the annual
demand is D items and that the firm orders Q of these at regular intervals during the year.
For example, if the demand is 12 000, the firm might place an order for 1000 items each
month. Alternatively, the firm could increase the order size to 3000 items and thereby reduce
the number of orders to just 12 000/3000 = 4 times a year. In general, orders are made D/Q
times a year.
There are many costs incurred in inventory control (including opportunity costs), but
we shall consider just two: ordering costs and holding costs. We consider each of these
in turn.
Section 4.7 Further optimisation of economic functions 357
There will always be a cost of raising an order which includes a fixed administrative
charge, delivery charge and the cost of setting up a production run. If the cost of placing each
order is $R, then the total cost is found by multiplying R by the number of orders, D/Q, so
total annual order cost 5
DR
Q
The other major cost incurred is the stockholding cost. This includes the cost of storage,
insurance and interest lost on the capital used to buy these goods waiting to be sold. In the
simplest model we will assume that the level of stock falls at a constant rate throughout the
year and that the firm waits until the stock levels are zero before reordering. If we further
assume that the stock is replenished instantly, then the variation of stock levels during the
course of a year would be modelled by the pattern shown in Figure 4.28. In practice this
may not occur. There are likely to be seasonal variations in demand, so the gradients of each
section of the graph may vary from one cycle to the next, and demand may not even be linear.
Also, it is not good practice to wait until shelves are empty before reordering stock and, since
goods do not arrive instantly, firms build in a lead time to allow for this. However, as a first
model of the real situation, we will assume that stock levels follow the pattern shown in
Figure 4.28.
In each period the average number of items stored is Q/2, and, since this pattern is
replicated throughout the year, the average number of items held in store each year is Q/2. If
it costs $H to store one item for one year, then the total annual holding cost is
H3
Q
HQ
5
2
2
If C denotes the total cost per annum, then we can add the order and holding costs to deduce
that
C5
DR
HQ
1
Q
2
Figure 4.29 shows a graph of order cost, holding cost and total cost plotted against Q. As
expected, to cut down on the order costs it is best to make Q as large as possible so that you
don’t have to incur the overheads for ordering the goods. On the other hand, to minimise
holding costs you want to make Q as small as possible. These conflicting interests lead to an
overall single minimum as shown by the graph of C in Figure 4.29.
Stock
Q
Time
Figure 4.28
358 Chapter 4 Differentiation
Cost, C
total cost
holding cost
order cost
Order quantity, Q
Figure 4.29
At a stationary point, dC 5 0, so we need to differentiate
dQ
C 5 (DR)Q 21 1
HQ
2
with respect to Q. The remaining letters, D, R and H, are all constants so
dC
H 2DR
H
5 2(DR)Q22 1
5
1
dQ
2
Q2
2
The stationary point can be found by solving
H
DR
5 2
2
Q
for Q:
Q2 5
so
Q5
2DR
H
2DR
H
This value of Q is called the economic order quantity (EOQ). It provides a formula for the
optimal order size in terms of annual demand, D, cost of placing each order, R, and annual
cost of holding an item in stock, H. If the order exceeds EOQ, then you order less frequently
and, although the ordering costs decrease, the holding costs rise more than this. On the other
hand, if the order is below the EOQ, then you order more frequently and, although the holding costs decrease, this is offset by higher ordering costs. The graph shown in Figure 4.29
shows clearly that this stationary point is a minimum, but it is possible to show this by inves­
tigating the sign of the second-order derivative, d2C/dQ2 (see Question 5 in Exercise 4.7*).
Section 4.7 Further optimisation of economic functions 359
Key Term
Economic order quantity The quantity of a product that should be ordered so as to minimise the
total cost that includes ordering costs and holding costs.
Exercise 4.7*
1. A firm’s demand function is
P = aQ + b (a < 0, b > 0)
Fixed costs are c and variable costs per unit are d.
(a) Write down general expressions for TR and TC.
(b) By differentiating the expressions in part (a), deduce MR and MC.
(c) Use your answers to (b) to show that profit, π, is maximised when
Q5
d2b
2a
2. (a) In Section 4.5 the following relationship between marginal revenue, MR, and price
elasticity of demand, E, was derived:
MR 5 P 1 1
1
E
Use this result to show that at the point of maximum total revenue, E = −1.
(b) Verify the result of part (a) for the demand function
2P + 3Q = 60
3. The demand functions for a firm’s domestic and foreign markets are
P1 = 50 − 5Q1
P2 = 30 − 4Q2
and the total cost function is
TC = 10 + 10Q
where Q = Q1 + Q2. Determine the prices needed to maximise profit
(a) with price discrimination;
(b) without price discrimination.
Compare the profits obtained in parts (a) and (b).
4. Show that if the marginal cost curve cuts the marginal revenue curve from above, then
profit is a minimum.
5. (a) Find an expression for the second-order derivative, d2C/dQ2, of the cost function
C5
DR
HQ
1
Q
2
and hence show that the economic order quantity, Q 5
(b) Obtain a simplified expression for the minimum cost.
2DR
, is a minimum point.
H
➜
360 Chapter 4 Differentiation
6. (a) The annual demand of a good is 2000 units, the fixed cost of placing an order is $40
and the annual cost of storing an item is $100. Assuming that the same order is
placed at regular intervals throughout the year, and that the firm waits for stock
levels to reduce to zero before ordering new stock, work out how many items should
be ordered each time to minimise total costs. What is the minimum total cost?
(b) Repeat part (a) if the annual cost of storage costs falls to $64.
(c) Repeat part (a) if the fixed cost of placing an order rises to $160.
(d) What effect do changes in order costs and holding costs have on the minimum total
cost?
7. If the total cost function TC = aQ2 + bQ + c, write down an expression for the average
cost function, AC. Show that AC is a minimum when Q 5
c
and find the corresponda
ing value of AC.
8. (a) Show that, at a stationary point of an average cost function, average cost equals
marginal cost.
(b) Show that if the marginal cost curve cuts the average cost curve from below, then
average cost is a minimum.
9. In a competitive market the equilibrium price, P, and quantity, Q, are found by setting
QS = QD = Q in the supply and demand equations
P = aQS + b (a > 0, b > 0)
P = −cQD + d (c > 0, d > 0)
If the government levies an excise tax, t, per unit, show that
Q5
d2b2t
a1c
Deduce that the government’s tax revenue, T = tQ, is maximised by taking
t5
d2b
2
Section 4.8
The derivative of the exponential and
natural logarithm functions
Objectives
At the end of this section you should be able to:
●
Differentiate the exponential function.
●
Differentiate the natural logarithm function.
●
Use the chain, product and quotient rules to differentiate combinations of these functions.
●
Appreciate the use of the exponential function in economic modelling.
In this section we investigate the derived functions associated with the exponential and
natural logarithm functions, ex and ln x. The approach that we adopt is similar to that used
in Section 4.1. The derivative of a function determines the slope of the graph of a function.
Consequently, to discover how to differentiate an unfamiliar function, we first produce an
accurate sketch and then measure the slopes of the tangents at selected points.
Advice
The functions ex and ln x were first introduced in Section 2.4. You might find it useful to remind
yourself how these functions are defined before working through the rest of the current section.
Figure 4.30 shows a sketch of the exponential function, ex, based on the table of values:
x
–2.0
–1.5
–1.0
–0.5
0.0
0.5
1.0
1.5
f(x)
0.14
0.22
0.37
0.61
1.00
1.65
2.72
4.48
From the graph we see that the slopes of the tangents at x = −1, x = 0 and x = 1 are
0.20
5 0.4
0.50
0.50
f9(0) 5
5 1.0
0.50
1.35
5 2.7
f9(1) 5
0.50
f9(21) 5
These results are obtained by measurement and so are quoted to only one decimal place. We
cannot really expect to achieve any greater accuracy using this approach.
362 Chapter 4 Differentiation
y
5
y 5 ex
4
3
Dy 5 1.35
2
1
Dx 5 0.50
Dy 5 0.50
Dx 5 0.50
Dy 5 0.20
Dx 5 0.50
22
21
0
1
2
x
Figure 4.30
The values of x, f(x) and f ′(x) are summarised in the following table. The values of
f(x) are rounded to one decimal place in order to compare with the graphical estimates
of f ′(x).
x
–1
0
1
f(x)
f ′(x)
0.4
0.4
1.0
1.0
2.7
2.7
Notice that the values of f (x) and f ′(x) are identical to within the accuracy quoted.
These results suggest that the slope of the graph at each point is the same as the function
value at that point: that is, ex differentiates to itself. Symbolically,
if f (x) = ex then f ′(x) = ex
or, equivalently,
if y 5 ex then
dy
5 ex
dx
Section 4.8 The derivative of the exponential and natural logarithm functions 363
Practice Problem
1. Use your calculator to complete the following table of function values and hence sketch
an accurate graph of f (x) = ln x:
x
0.50
1.00
f(x)
1.50
2.00
2.50
3.00
0.41
3.50
4.00
1.25
Draw the tangents to the graph at x = 1, 2 and 3. Hence estimate the values of f ′(1),
f ′(2) and f ′(3). Suggest a general formula for the derived function f ′(x).
[Hint: for the last part you may find it helpful to rewrite your estimates of f ′(x) as
simple fractions.]
In fact, it is possible to prove that, for any value of the constant m,
if
y 5 emx then
dy
5 memx
dx
and
if
y 5 In mx then
dy
5 1x
dx
In particular, we see by setting m = 1 that
ex differentiates to ex
and that
ln x differentiates to
1
x
which agree with our practical investigations.
Example
Differentiate
(a) y = e2x
(b) y = e−7x
(c) y = ln 5x (x > 0)
(d) y = ln 559x (x > 0)
Solution
(a) Setting m = 2 in the general formula shows that
if
y 5 e2x
then
dy
5 2e2x
dx
Notice that when exponential functions are differentiated, the power itself does not
change. All that happens is that the coefficient of x comes down to the front.
➜
364 Chapter 4 Differentiation
(b) Setting m = −7 in the general formula shows that
if y 5 e 27x then
dy
5 27e 27x
dx
(c) Setting m = 5 in the general formula shows that
if
y 5 ln 5x
then
dy
1
5
dx
x
Notice the restriction x > 0 stated in the question. This is needed to ensure that we do
not attempt to take the logarithm of a negative number, which is impossible.
(d) Setting m = 559 in the general formula shows that
if
y 5 ln 559x
then
dy
1
5
dx
x
Notice that we get the same answer as part (c). The derivative of the natural logarithm
function does not depend on the coefficient of x. This fact may seem rather strange,
but it is easily accounted for. The first rule of logarithms shows that ln 559x is the same
as
ln 559 + ln x
The first term is merely a constant and so differentiates to zero, and the second term
differentiates to 1/x.
Practice Problem
2. Differentiate
(a) y = e3x (b) y = e−x (c) y = ln 3x (x > 0) (d) y = ln 51 234x (x > 0)
The chain rule can be used to explain what happens to the m when differentiating emx. The
outer function is the exponential, which differentiates to itself, and the inner function is mx,
which differentiates to m. Hence, by the chain rule,
if
y 5 emx
then
dy
5 emx 3 m 5 memx
dx
Similarly, noting that the natural logarithm function differentiates to the reciprocal function,
if
y 5 ln mx
then
dy
1
1
3m5
5
mx
x
dx
The chain, product and quotient rules can be used to differentiate more complicated
functions involving ex and ln x.
Section 4.8 The derivative of the exponential and natural logarithm functions 365
Example
Differentiate
(a) y = x3e2x (b) y = ln(x2 + 2x + 1) (c) y 5
e3x
x 12
2
Solution
(a) The function x3e2x involves the product of two simpler functions, x3 and e2x, so we need
to use the product rule to differentiate it. Putting
u = x3
and v = e2x
gives
du
5 3x2 and
dx
dv
5 2e2x
dx
By the product rule,
dy
dv
du
5u
1v
5 x3 2e2x 1 e2x 3x2 5 2x3e2x 1 3x2e2x
dx
dx
dx
There is a common factor of x2e2x, which goes into the first term 2x times and into the
second term 3 times. Hence
dy
5 x2e2x (2x 1 3)
dx
(b) The expression ln(x2 + 2x + 1) can be regarded as a function of a function, so we can
use the chain rule to differentiate it. We first differentiate the outer log function to get
1
x2 1 2x 1 1
and then multiply by the derivative of the inner function, x2 + 2x + 1, which is 2x + 2.
Hence
dy
2x 1 2
5 2
dx
x 1 2x 1 1
(c) The function
e3x
x2 1 2
is the quotient of the simpler functions
u = e3x and v = x2 + 2
for which
du
5 3e3x and
dx
dv
5 2x
dx
By the quotient rule,
du
dv
v
2u
e3x(3x2 2 2x 1 6)
dy
e3x 3(x2 1 2) 2 2x
dx
dx (x2 1 2)(3e3x) 2 e3x(2x)
5
5
5
5
dx
v2
(x2 1 2)2
(x2 1 2)2
(x2 1 2)2
366
chaptEr 4 DIFFEREnTIATIon
practice problem
3. Differentiate
(a) y = x4 ln x
(b) y = ex
2
(c) y 5
lnx
x12
Notice the effect that the chain rule has on the derivative of an exponential function such as
2
y = ex which you differentiated in part (b) of Practice Problem 3. In general:
If y = eu(x) then
dy
5 eu(x) 3 u9 (x)
dx
All you have to do is to write down the original function and then multiply by the derivative
2
2
of the exponent. To differentiate y = ex , first write down ex and then multiply it by
2
d 2
x 5 2x to get 2xex .
dx
advice
If you ever need to differentiate a function of the form:
ln (an inner function involving products, quotients or powers of x)
then it is usually quicker to use the rules of logs to expand the expression before you begin. The three
rules are
Rule 1
ln(x × y) = ln x + ln y
Rule 2
ln(x ÷ y) = ln x − ln y
Rule 3
ln xm = m ln x
The following example shows how to apply this ‘trick’ in practice.
Example
Differentiate
(a) y = ln(x(x + 1)4)
(b) y 5 ln
x
(x 1 5)
Solution
(a) From rule 1
ln(x(x + 1)4) = ln x + ln(x + 1)4
which can be simplified further using rule 3 to give
y = ln x + 4 ln(x + 1)
Differentiation of this new expression is simple. We see immediately that
dy
1
4
5 1
dx
x
x11
Section 4.8 The derivative of the exponential and natural logarithm functions 367
If desired, the final answer can be put over a common denominator
(x 1 1) 1 4x
1
4
5x 1 1
1
5
5
x
x11
x(x 1 1)
x(x 1 1)
(b) The quickest way to differentiate
y 5 ln
x
(x 1 5)
is to expand first to get
y 5 ln x 2 ln(x 1 5)1/2
(rule 2)
1
ln(x 1 5)
2
(rule 3)
5 ln x 2
Again this expression is easy to differentiate:
dy 1
1
5 2
dx
x 2(x 1 5)
If desired, this can be written as a single fraction:
1
1
2(x 1 5) 2 x
x 1 10
2
5
5
x
2(x 1 5)
2x(x 1 5)
2x(x 1 5)
Practice Problem
4. Differentiate the following functions by first expanding each expression using the rules
of logs:
(a) y = ln(x3(x + 2)4)
(b) y 5 ln
x2
2x 1 3
Exponential and natural logarithm functions provide good mathematical models in many
areas of economics, and we conclude this chapter with some illustrative examples.
Example
A firm’s short-run production function is given by
Q = L2e−0.01L
Find the value of L that maximises the average product of labour.
Solution
The average product of labour is given by
APL 5
Q
L2e20.01L
5 Le20.01L
5
L
L
To maximise this function we adopt the strategy described in Section 4.6.
➜
368 Chapter 4 Differentiation
Step 1
At a stationary point
d(APL)
50
dL
To differentiate Le−0.01L, we use the product rule. If
u = L and v = e−0.01L
then
du
5 1 and
dL
dv
5 20.01e20.01L
dL
emx differentiates
to memx
By the product rule,
d(APL)
dv
du
5u
1v
5 L(20.01e20.01L) 1 e20.01L 5 (1 2 0.01L)e20.01L
dL
dL
dL
We know that a negative exponential is never equal to zero. (Although e−0.01L gets ever
closer to zero as L increases, it never actually reaches it for finite values of L.) Hence the
only way that
(1 − 0.01L)e−0.01L
can equal zero is when
1 − 0.01L = 0
which has the solution L = 100.
Step 2
To show that this is a maximum we need to differentiate a second time. To do this we
apply the product rule to
(1 − 0.01L)e−0.01L
taking
u = 1 − 0.01L and
v = e−0.01L
for which
du
dv
5 20.01 and
5 20.01e20.01L
dL
dL
Hence
d2(APL)
dv
du
5u
1v
5 (1 2 0.01L)(20.01e20.01L) 1 e20.01L(20.01) 5 (20.02 1 0.0001L)e20.01L
dL2
dL
dL
Finally, putting L = 100 into this gives
d2(APL)
5 20.0037
dL2
The fact that this is negative shows that the stationary point, L = 100, is indeed a maximum.
Section 4.8 The derivative of the exponential and natural logarithm functions 369
Practice Problem
5. The demand function of a good is given by
Q = 1000e−0.2P
If fixed costs are 100 and the variable costs are 2 per unit, show that the profit function
is given by
π = 1000Pe−0.2P − 2000e−0.2P − 100
Find the price needed to maximise profit.
Example
A firm estimates that the total revenue received from the sale of Q goods is given by
TR = ln(1 + 1000Q2)
Calculate the marginal revenue when Q = 10.
Solution
The marginal revenue function is obtained by differentiating the total revenue function.
To differentiate ln(1 + 1000Q2), we use the chain rule. We first differentiate the outer log
function to get
1
1 1 1000Q2
natural logs
differentiate to
reciprocals
and then multiply by the derivative of the inner function, 1 + 1000Q2, to get 2000Q. Hence
MR 5
d(TR)
2000Q
5
dQ
1 1 1000Q2
At Q = 10,
MR 5
2000(10)
5 0.2
1 1 1000(10)2
Practice Problem
6. If the demand equation is
P = 200 − 40 ln(Q + 1)
calculate the price elasticity of demand when Q = 20.
370 Chapter 4 Differentiation
Exercise 4.8
1. Write down the derivative of
(a) y = e6x (b) y = e−342x (c) y = 2e−x + 4ex (d) y = 10e4x – 2x2 + 7
2. If $4000 is saved in an account offering a return of 4% compounded continuously, the
future value, S, after t years is given by
S = 4000e0.04t
(1) Calculate the value of S when
(a) t = 5 (b) t = 5.01
and hence estimate the rate of growth at t = 5. Round your answers to two decimal
places.
(2) Write down an expression for
growth after five years.
dS
and hence find the exact value of the rate of
dt
3. Write down the derivative of
(a) y = ln(3x) (x > 0)
(b) y = ln(−13x) (x < 0)
4. Use the chain rule to differentiate
3
(a) y 5 ex
(b) y = ln(x4 + 3x2)
5. Use the product rule to differentiate
(a) y = x4e2x
(b) y = x ln x
6. Use the quotient rule to differentiate
(a) y 5
e4x
x 12
2
(b) y 5
ex
ln x
7. Find and classify the stationary points of
(a) y = xe−x
(b) y = ln x − x
Hence sketch their graphs.
8. Since the beginning of the year, weekly sales of a luxury good are found to have
decreased exponentially. After t weeks, sales can be modelled by 3000e−0.02t.
(a) Work out the weekly sales when t = 12 and t = 13 and hence find the decrease in
sales during this time.
(b) Use differentiation to work out the rate of decrease in sales after 12 weeks and
compare this with your answer to part (a).
9. Find the output needed to maximise profit given that the total cost and total revenue
functions are
TC = 2Q and TR = 100 ln(Q + 1)
respectively.
10. If a firm’s production function is given by
Q = 700Le−0.02L
find the value of L that maximises output.
Section 4.8 The derivative of the exponential and natural logarithm functions 371
11. The demand function of a good is given by
P = 100e−0.1Q
Show that demand is unit elastic when Q = 10.
Exercise 4.8*
1. Differentiate:
(a) y = e2x − 3e−4x (b) xe4x (c)
(f)
emx
xn
(g)
(ax 1 b)n
ln x
(h)
e2x
(d) x m ln x (e) x(ln x − 1)
x2
e ax
(ln bx) n
(i)
ex 2 1
ex 1 1
2. Use the rules of logarithms to expand each of the following functions. Hence find their
derivatives.
(a) y 5 ln
x
(b) y 5 ln(x (3x 2 1)) (c) y 5 ln
x11
x11
x21
dy
4 y.
dt
(a) Use this definition to find the growth rate of the variable, y = Aekt.
3. The growth rate of an economic variable, y, is defined to be
(b) The gross domestic product, GDP, and the size of the population, N, of a country
grow exponentially, so that after t years, GDP = Aeat and N = Bebt.
(i) State the growth rates for GDP and N.
(ii) Show that the GDP per capita also grows exponentially and write down its
growth rate.
4. Differentiate the following functions with respect to x, simplifying your answers as far
as possible:
2
(a) y = x4e−2x (b) y 5 ln
x
(x 1 1)2
5. Find and classify the stationary points of
(a) y = xeax (b) y = ln(ax 2 + bx)
where a < 0.
6. (a) Use the quotient rule to show that the derivative of the function
y5
2x 1 1
4x 1 3
is given by
4(x 1 1)
(4x 1 3) 4x 1 3
➜
372 Chapter 4 Differentiation
(b) Use the chain rule to differentiate the function
y 5 ln
2x 1 1
4x 1 3
(c) Confirm that your answer to part (b) is correct by first expanding
ln
2x 1 1
4x 1 3
using the rules of logs and then differentiating.
7. A firm’s short-run production function is given by
Q = L3e−0.02L
Find the value of L that maximises the average product of labour.
8. Find an expression for the price elasticity of demand for the demand curve
P = 500 − 75 ln(2Q + 1)
9. Find an expression for the marginal revenue for each of the following demand curves:
2
Q
2Q
(a) P 5 e (b) P 5 ln
3Q 1 1
Q2
10. The demand function of a good is given by Q = 4000e−0.01P
(a) Find an expression, in terms of P, for the elasticity of demand and hence determine
the range of values of P when the demand is inelastic.
(b) Find the price which maximises total revenue.
11. If the total cost function is given by TC 5 20 QeQ/4, find the value of Q which minimises
average cost.
12. The logistic model of growth takes the general form,
y5
k
1 1 be2at
where k, a and b are positive constants.
(a) Find an expression for dy/dt and deduce that the gradient is positive.
(b) Find an expression for d2y/dt2 and deduce that the graph is convex when t < (ln b)/a
and concave when t > (ln b)/a.
(c) State the coordinates of the point where the graph intercepts the y axis, and describe
the behaviour of the graph as t S ∞.
(d) Sketch a graph of this logistic function.
13. An art collector owns a painting which is currently valued at $2 million. After t years it
is expected that the painting will be worth V million dollars where V 5 2e t .
(a) If the interest rate is 10% compounded continuously, show that after t years the
present value of the painting is given by PV 5 2e t20.1t.
(b) The collector decides to sell the painting after T years where T is chosen to maximise
PV. Work out the value of T.
Formal mathematics
In more advanced books on mathematics, the derivative is defined via the concept of a limit
and is usually written in symbols as
Dy
dy
5 lim
dx Dx ` Dx
Look at Figure 4.31. Points A and B both lie on the curve y = f(x) and their x and y coordinates differ by Δx and Δy, respectively. A line AB which joins two points on the curve is known
as a chord, and it has slope Δy/Δx.
Now look at Figure 4.32, which shows a variety of chords, AB1, AB2, AB3, . . . , corresponding
to smaller and smaller ‘widths’ Δx. As the right-hand end points, B1, B2, B3, . . . , get closer
to A, the ‘width’, Δx, tends to zero. More significantly, the slope of the chord gets closer to
that of the tangent at A. We describe this by saying that in the limit, as Δx tends to zero, the
slope of the chord, Δy/Δx, is equal to that of the tangent. This limit is written
lim
Dx
0
Dy
Dx
We deduce that the formal definition
Dy
dy
5 lim
dx Dx ` Dx
coincides with the idea that dy/dx represents the slope of the tangent, which is the approach
adopted in this text.
y
y 5 f(x)
B
chord
Dy
A
Dx
x
Figure 4.31
374 Chapter 4 Differentiation
y
y 5 f(x)
B1
B2
chords
B3
A
B4
tangent
x
Figure 4.32
Throughout this chapter we have assumed that all functions can be differentiated. The
function must at least be continuous at a point before we can even think about trying to draw
a tangent at any point. For the function sketched in Figure 4.33, tangents could be drawn at
every point on the graph except at x = 2, where the graph is discontinuous. We say that this
function is not differentiable at x = 2.
2
Figure 4.33
Formal mathematics 375
y
y 5 2x
y5x
gradient 5 21
here
gradient 5 1
here
x
Function not
differentiable
at x 5 0
Figure 4.34
Even if a function is continuous everywhere, it still might not be possible to draw a tangent
unless the curve is smooth. A classic example of such a function is the modulus function,
f(x) = u x u which is defined as
x 5
2x, if x , 0
x, if x $ 0
The graph of the modulus function is sketched in Figure 4.34 and has a sharp corner at
x = 0, making it impossible to draw a tangent there.
To the left of the origin the graph has a constant slope of −1, and to the right of the origin
the graph has a constant slope of +1. However, at x = 0 we cannot draw a tangent, so the
modulus function is differentiable at all values of x except for x = 0. This can also be seen from
the definition,
f 9(x) 5 lim
Dx
0
Dy
Dx
As Δx S 0 through positive values (i.e. from the right) the limit is +1, whereas as Δx S 0
through negative values (i.e. from the left) the limit is −1. The limits can both be found, but
the fact that they are different values prevents us from finding the derivative at x = 0.
Multiple choice questions
Question 1
A graph of a function, y = f(x), is shown in the diagram. By drawing a tangent, estimate
f ′(6).
y
5
2
4
6
8
x
25
210
A. 0
B. −2
C. 2
D. 4
E. −4
Question 2
Work out the gradient of the curve y 5 x2 2 2x 1 4 x 2
A. 8
B. 7
C. 6
16
at x = 4.
x
D. 5
E. 4
Question 3
If the total revenue and total cost functions are TR = 36Q − Q2 and TC = 10 + 3Q, find
the maximum profit.
A. 272.25
B. 262.25
C. 282.25
D. 292.25
E. 302.25
Question 4
The total revenue function is TR = 120Q − 5Q2 and Q = 5. Use differentiation to estimate
the change in TR brought about by a 0.25 increase in Q.
A. 17.50
B. 17.81
C. 23.75
D. 26.24
E. 14.75
Multiple choice questions 377
Question 5
A cubic function is given by y = 2x3 − 5x2 + 13x − 40. Is this function increasing or
decreasing, and is it convex or concave, at x = 0.5?
A. Increasing and concave
B. Decreasing and concave
C. Increasing and convex
D. Decreasing and convex
E. Indeterminate
Question 6
Graphs of marginal revenue and marginal cost functions based on linear models for the
demand and total cost functions of a good are shown in the diagram. The quantity is currently 20 units. Describe the qualitative effect on total revenue, total cost and profit if the
quantity increases to 21 units.
y
50
40
30
MR
20
MC
10
20
210
40 60 80 100
Quantity
x
A. TR decreases, TC decreases, π decreases
B. TR increases, TC increases, π decreases
C. TR decreases, TC decreases, π increases
D. TR decreases, TC increases, π decreases
E. TR increases, TC increases, π increases
Question 7
Find the y-coordinate of the (local) maximum point on the graph of the cubic function
f(x) = −x3 + 21x2 − 135x + 300.
A. 25
B. 39
C. 57
D. 5
E. 9
Question 8
If the supply function Q = 10 + 0.1P + aP2 is unit elastic at P = 10, find the value of a.
A. 0.1
B. 0.2
C. 0.3
D. 0.4
E. 0.5
378 Chapter 4 Differentiation
Question 9
If the demand function is P = (40 − 2Q)2 on 0 ≤ Q ≤ 20, find the marginal revenue at
Q = 5.
B. −120
A. 120
C. −60
D. −1140
E. 300
Question 10
Calculate the MPC at Y = 8, if the consumption function is C 5
A. 0.92
B. 0.14
C. 0.42
D. 0.86
10 1 Y 2
.
21Y
E. 0.74
Question 11
The weekly sales S(t) of a new product follow the logistic model: S(t) 5
10 000
1 1 5e20.02t
Find the rate of increase in sales after 10 weeks.
A. 31.6
B. 29.7
C. 38.5
D. 42.3
E. 45.9
Question 12
Find an expression for the marginal revenue function in terms of Q when the demand
function is Q 5
100 2 P
2P
100
(2Q 1 1)2
B. 2
A.
200
(2Q 1 1)2
C.
200
(2Q 1 1)2
D.
100 1 Q
(2Q 1 1)2
E.
100 2 Q
(2Q 1 1)2
Question 13
Find the gradient of the curve y = x(3x + 2)3 at x = 2.
A. 896
B. 1664
C. 1152
D. 384
Question 14
If f ″(x) = 0 and f ′(x) ≠ 0 for all x, what type of function is f(x)?
A. Zero
B. Constant
C. Linear
D. Quadratic
E. Cubic
E. 128
Multiple choice questions 379
Question 15
In the simplest model of inventory control, which of the graphs shown in the diagram
represent total cost, holding cost and order cost?
B
A
C
Q
A. A is total cost, B is order cost and C is holding cost
B. A is total cost, B is holding cost and C is order cost
C. A is holding cost, B is total cost and C is order cost
D. A is holding cost, B is order cost and C is total cost
E. A is order cost, B is holding cost and C is total cost
Question 16
If the demand function is P = 40 − 5Q, find the values of P for which demand is inelastic.
A. P > 8
B. P > 20
C. P < 8
D. P < 20
E. P > 4
Question 17
The supply function of a good is given by P = 4 + 2Q + 0.1Q2. Calculate the arc elasticity
of supply as Q increases from 10 to 12.
A. 1.04
B. 0.07
C. 1.21
D. 0.34
E. 0.83
Question 18
If the total cost of producing Q goods is given by TC = Q2 + 10Q + 9, find the level of
output which minimises average cost.
A. 2
B. 3
C. 4
D. 5
E. 6
380 Chapter 4 Differentiation
Question 19
What type of economic problem is solved by equating marginal product of labour and
average product of labour?
A. Maximise output
B. Maximise labour productivity
C. Maximise profit
D. Minimise marginal product of labour
E. Minimise average product of labour
Question 20
Weekly sales of a good have decreased exponentially. After t weeks, sales can be modelled
by 4500e−0.05t. Use differentiation to find the rate of decrease after 20 weeks.
A. 225
B. 166
C. 4500
D. 450
E. 83
Question 21
Find
x
dy
when y 5 ln
, expressing your answer as a single fraction.
x14
dx
A.
x 18
2x(x 1 4)
B.
3x 1 4
2x(x 1 4)
C.
42x
2x(x 1 4)
D.
22x
2x(x 1 4)
E.
3x 1 8
2x(x 1 4)
Question 22
The price and demand of a good are currently $100 and 400, respectively. After a price
increase of $20, the demand falls by 50. Calculate the arc elasticity of demand.
3
4
5
2
8
6
2
7
11
2
15
2
2
3
A. 2
B.
C.
D.
E.
Multiple choice questions 381
Question 23
The demand function of a good is P = −Q2 − 4Q + 120. Calculate the price elasticity of
demand when P = 75.
15
A. 2
14
B. 2
9
7
C. 2
9
8
11
D. 2
15
15
E. 2
11
Question 24
The graph of a derived function, f ′(x) is given by:
y
1
2
3
4
5
6 x
Which one of the graphs below shows the original function, f(x)?
y
y
1
2
3
4
5
6 x
A.
2
3
4
5
6 x
1
2
3
4
5
6 x
B.
y
y
1
2
3
4
5
6 x
C.
D.
y
1
E.
1
2
3
4
5
6 x
Examination questions
Question 1
Find an expression, in terms of Q, for the marginal revenue when the demand function is
Q + 4P = 70. Hence
(a) estimate the change in total revenue when Q decreases from 40 to 39.75;
(b) find the value of Q at which the total revenue is a maximum.
Question 2
A cubic function is given by f(x) = (x + 3)(x − 1)(9 − 2x).
(a) By multiplying out the brackets, show that f(x) = −2x3 + 5x2 + 24x − 27.
(b) Find an expression for the first-order derivative and hence find the coordinates of the
stationary points.
(c) Find an expression for the second-order derivative and hence classify the stationary
points found in part (b).
(d) Sketch a graph of f(x) showing clearly on your sketch where the curve crosses the coor-
dinate axes.
Question 3
(a) Find the gradient of each of the following curves at x = 9:
(i) y = 5x2 (ii) y 5
27
(iii) y 5 6 x
x
(b) If the supply function of a good is Q = 0.01P2 + P + 30, calculate the price elasticity
of supply
(i) averaged along an arc between P = 20 and P = 30;
(ii) at the point P = 25.
Question 4
(a) Consider the function f (x) = 2x3 + 7x + 1.
(i) Work out the values of f(2) and f(2.1), and hence calculate the gradient of the
straight line joining (2, f(2)) and (2.1, f (2.1)).
(ii) Work out the value of f ′(2).
(iii) By using a sketch graph, explain why your answers to parts (i) and (ii) are very
similar but are not equal.
Examination questions 383
(b) The average cost function of a good is given by AC 5 3Q 1 8 1
15
Q
(i) Find an expression for the marginal cost function.
(ii) If the current output is 30, use your answer to part (i) to estimate the effect on TC
of a 0.5-unit increase in Q, and compare this with the actual change in TC.
Question 5
The demand and supply functions of a good are given by P + 2QD = 13 and P − QS = 4,
respectively.
(a) Determine the market equilibrium price.
(b) Find the elasticity of supply at this price.
(c) Estimate the percentage increase in price needed to increase supply by 2% of its equi-
librium value. Is supply elastic, inelastic or unit elastic at market equilibrium?
Question 6
(a) Differentiate
(i) y = 6e2x (ii) y = ln(x2 + 1)
(b) Evaluate the first- and second-order derivatives of the cubic y = x3 + 2x2 − 6x + 7 at x = 2,
and hence state whether the graph is
(i) increasing or decreasing at this point;
(ii) convex or concave at this point.
Question 7
(a) Find the gradient of each of the following curves at x = 1:
(i) y = x(1 + x)4
(ii) y = (x2 + 6)3
2x
x 11
x 11
(iv) y 5 ln
x13
(iii) y 5
(b) A firm’s short-run production function is given by Q = Le−0.02L. Find the marginal product of labour at L = 50, and give a brief interpretation of this value.
Question 8
A company monitors its daily profit during the first five years of trading. The variation of
profit over time can be modelled approximately by the cubic
π = t3 − 9t2 + 24t − 16
(0 ≤ t ≤ 5)
where t denotes the time in years since the launch of the company.
384 Chapter 4 Differentiation
(a) Verify that the firm makes a loss initially and that it breaks even after the first year.
(b) Solve the equation, dπ/dt = 0 to find the coordinates of the stationary points on the
curve.
(c) By examining the sign of the second-order derivative d2π/dt2 at each stationary point,
determine their nature.
(d) Sketch a graph of π against t.
(e) Identify the times at which the profit function is increasing and decreasing.
(f) Identify the times at which the graph is concave and convex.
(g) Summarise, in non-mathematical terms, how the firm’s profit has changed during the
first five years of trading.
Question 9
A firm’s short-run production function is given by Q = 10L + 6L2 − 0.1L3.
(a) Write down expressions for the marginal product of labour and average product of
labour.
(b) Determine the value of L which maximises the average product of labour.
(c) Use calculus to show that the law of diminishing returns holds.
Question 10
(a) Find an expression, simplified as far as possible, for marginal revenue for the demand
function
P5
400 2 6Q
(b) Find an expression for the marginal propensity to consume for the consumption
function C 5
MPS 5
200 1 Y 2
and deduce that the marginal propensity to save is given by
10 1 Y
300
.
(10 1 Y )2
Question 11
The supply and demand functions of a good are given by
P = aQS + b
P = −cQD + d
where P, QS and QD denote the price, quantity supplied and quantity demanded, respectively, and where a, b, c and d are positive constants.
The government imposes a fixed tax, t, per unit.
(a) By drawing sketch graphs on the same diagram, explain why the condition d − b > 0
is required for the pre-tax market equilibrium to exist.
(b) Find expressions for the equilibrium price and quantity both before and after the tax
is imposed.
Examination questions 385
(c) Write down the ratio of the tax that is paid by the supplier and consumer, respectively.
(d) Find the value of t which optimises government tax revenue and show mathematically
that this is a maximum.
Question 12
A monopolistic firm produces goods in a market where the demand function is P = 43 − 0.3Q
and the corresponding total cost function is TC = 0.01Q3 − 0.4Q2 + 3Q.
(a) What can you say about the fixed costs of this firm?
(b) Find the (non-zero) output for which average cost is equal to marginal cost, and
explain the significance of this value.
(c) Find the output which maximises profit and calculate the price elasticity of demand at
this point.
Question 13
A firm’s demand function is given by P 5 a 1 b Q where P and Q denote the price and
quantity demanded, respectively, and a and b are parameters (a > 0, b < 0).
(a) Obtain a general expression for the price elasticity of demand in terms of P and deduce
that demand is unit elastic when P = a/3.
(b) Write down an expression for total revenue, TR, and hence find the value of Q at the
stationary point of this function. By considering the second-order derivative of TR,
show that your stationary point is a maximum.
Question 14
(a) By applying the product rule to the expression TR = PQ, prove that marginal revenue,
MR, satisfies
MR 5 P 1 1
1
E
where E is the price elasticity of demand.
Hence explain why the total revenue is a decreasing function in regions where the
demand is inelastic.
(b) A building supplier is allowed to charge different prices to its trade and individual
customers. The corresponding demand equations are given by
Q1 = 200 − 2P1
Q2 = 250 − P2
respectively. The total cost function is TC = 100 + 20Q where Q = Q1 + Q2.
Find the prices that the firm should charge its trade and individual customers to maximise total profit.
Calculate the elasticities at these prices and verify that the firm charges the higher
price in the market where the magnitude of the elasticity of demand is lower. Use the
result of part (a) to explain why this is always true.
386 Chapter 4 Differentiation
Question 15
The diagram provides a model for the variation of stock levels of a good during the course
of a year where:
Stock
D = annual demand
Q = size of each order made at regular intervals during the year
R = cost of raising each order
H = cost of storing an item for a year.
U
Time
The diagram shows that stock levels fall at a constant rate and that these are instantly
replenished when the level falls to U units.
(a) Explain why the annual holding cost is given by AHC =
(b) Explain why the annual order cost is AOC =
1
Q 1 U H.
2
DR
.
Q
(c) Assuming that the total cost, C = AHC + AOC, show that C is minimised when
2DR
. Use second-order derivatives to show that your stationary point is a
Q5
H
minimum.
Question 16
(a) The supply function of a good is given by P = Q2 + 2Q + 8. Find the price elasticity of
supply when P = 128 and hence estimate the percentage change in supply due to a 5%
increase in price. Is supply elastic, inelastic or unit elastic at this price?
(b) The demand function of the same good is P = aQ + b. Find the values of a and b if it
is known that the market equilibrium price is 56 and that demand is unit elastic at
P = 40.
This page intentionally left blank
388  
Chapter 5
Partial Differentiation
This chapter continues the topic of calculus by describing how to differentiate functions of
more than one variable. In many ways this chapter can be regarded as the climax of the whole
text. It is the summit of the mathematical mountain that we have been merrily climbing. Not
only are the associated mathematical ideas and techniques quite sophisticated, but also partial
differentiation provides a rich source of applications. In one sense there is no new material
presented here. If you know how to differentiate a function of one variable, then you also know
how to partially differentiate a function of several variables because the rules are the same.
Similarly, if you can optimise a function of one variable, then you need have no fear of unconstrained and constrained optimisation. Of course, if you cannot use the elementary rules of
differentiation or cannot find the maximum and minimum values of a function as described in
Chapter 4, then you really are fighting a lost cause. Under these circumstances, you are best
advised to omit this chapter entirely. There is no harm in doing this, because it does not form
the prerequisite for any of the later topics. However, you will miss out on one of the most elegant and useful branches of mathematics.
It is important that Sections 5.1 and 5.2 are read first, but the remaining four sections can be
studied in any order. Sections 5.1 and 5.2 follow the familiar pattern. We begin by looking at the
mathematical techniques and then use them to determine marginal functions and elasticities.
Section 5.3 describes the multiplier concept and completes the topic of national income determination which you studied in Chapter 1.
The final three sections are devoted to optimisation. For functions of several variables, optimisation
problems are split into two groups, unconstrained and constrained. Unconstrained problems,
tackled in Section 5.4, involve the maximisation and minimisation of functions in which the variables
are free to take any values whatsoever. In a constrained problem only certain combinations of the
variables are examined. For example, a firm might wish to minimise costs but is constrained by
the need to satisfy production quotas, or an individual might want to maximise utility but is subject to
a budgetary constraint, and so on. There are two ways of solving constrained problems: the
method of substitution and the method of Lagrange multipliers, described in Sections 5.5 and
5.6, respectively.
Section 5.1
Functions of several variables
Objectives
At the end of this section you should be able to:
●
Use the function notation, z = f(x, y).
●
Determine the first-order partial derivatives, fx and fy.
●
Determine the second-order partial derivatives, fxx, fxy, fyx and fyy.
●
Appreciate that, for most functions, fxy = fyx.
●
Use the small increments formula.
●
Perform implicit differentiation.
Most relationships in economics involve more than two variables. The demand for a good
depends not only on its own price but also on the price of substitutable and complementary
goods, incomes of consumers, advertising expenditure and so on. Likewise, the output from
a production process depends on a variety of inputs, including land, capital and labour. To
analyse general economic behaviour, we must extend the concept of a function, and particularly the differential calculus, to functions of several variables.
A function, f, of two variables is a rule that assigns to each incoming pair of numbers, (x, y),
a uniquely defined outgoing number, z. This is illustrated in Figure 5.1. The ‘black box’ f
performs some arithmetic operation on x and y to produce z. For example, the rule might
be ‘multiply the two numbers together and add twice the second number’. In symbols we
write this either as
f(x, y) = xy + 2y
or as
z = xy + 2y
In order to be able to evaluate the function, we have to specify the numerical values of both
x and y.
For example, substituting x = 3 and y = 4 gives
f(3,4) = 3 × 4 + 2 × 4 = 20
x
y
Figure 5.1
f
z
Section 5.1 Functions of several variables 391
and substituting x = 4 and y = 3 gives
f (4,3) = 4 × 3 + 2 × 3 = 18
Notice that, for this function, f (3,4) is not the same as f (4,3), so in general we must be careful
to write down the correct ordering of the variables.
We have used the labels x and y for the two incoming numbers (called the independent
variables) and z for the outgoing number (called the dependent variable). We could equally
well have written the above function as
y = x1x2 + 2x2
say, using x1 and x 2 to denote the independent variables and using y this time to denote the
dependent variable. The use of subscripts may seem rather cumbersome, but it provides an
obvious extension to functions of more than two variables. In general, a function of n variables can be written
y = f (x1, x2, . . . , xn)
Practice Problem
1. If
f (x, y) = 5x + xy2 − 10
and
g(x1, x2, x3) = x1 + x2 + x3
evaluate
(a) f (0, 0)
(b) f (1, 2)
(c) f(2, 1)
(d) g(5, 6, 10)
(e) g(0, 0, 0)
(f) g(10, 5, 6)
A function of one variable can be given a pictorial description using graphs, which help to
give an intuitive feel for its behaviour. Figure 5.2 shows the graph of a typical function
y = f(x)
in which the horizontal axis determines the incoming number, x, and the vertical axis determines the corresponding outgoing number, y. The height of the curve directly above any
point on the x axis represents the value of the function at this point.
y
y 5 f(x)
x
Figure 5.2
x
392 Chapter 5 Partial Differentiation
z
z 5 f(x, y)
y
(x, y)
x
Figure 5.3
An obvious question to ask is whether there is a pictorial representation of functions of
several variables. The answer is yes in the case of functions of two variables, although it is
not particularly easy to construct. A function
z = f (x, y)
can be thought of as a surface, rather like a mountain range, in three-dimensional space as
shown in Figure 5.3. If you visualise the incoming point with coordinates (x, y) as lying in
a horizontal plane, then the height of the surface, z, directly above it represents the value of
the function at this point. As you can probably imagine, it is not an easy task to sketch the
surface by hand from an equation such as
f (x, y) = xy3 + 4x
although three-dimensional graphics packages are available for most computers which can
produce such a plot.
It is impossible to provide any sort of graphical interpretation for functions of more than
two variables. For example, a function of, say, four variables would require five dimensions,
one for each of the incoming variables and a further one for the outgoing variable! In spite of
this setback we can still perform the task of differentiating functions of several variables and,
as we shall see in the remaining sections of this chapter, such derivatives play a vital role in
analysing economic behaviour.
Given a function of two variables,
z = f (x, y)
we can determine two first-order derivatives. The partial derivative of f with respect to x is
written as
−z
−x
or
−f
−x
or
fx
and is found by differentiating f with respect to x, with y held constant. Similarly, the partial
derivative of f with respect to y is written as
−z
−y
or
−f
−y
or
fy
Section 5.1 Functions of several variables 393
and is found by differentiating f with respect to y, with x held constant. We use curly ds in
the notation
−f
−x
read ‘partial dee f
by dee x’
to distinguish partial differentiation of functions of several variables from ordinary differ­en­
tiation of functions of one variable. The alternative notation, fx, is analogous to the f ′ notation for ordinary differentiation.
Example
Find the first-order partial derivatives of the functions
(a) f(x, y) = x2 + y3 (b) f (x, y) = x2y
Solution
(a) To differentiate the function
f (x, y) = x2 + y3
with respect to x, we work as follows. By the sum rule we know that we can differentiate each part separately and add. Now, when we differentiate x2 with respect to x, we
get 2x. However, when we differentiate y3 with respect to x, we get 0. To see this, note
from the definition of partial differentiation with respect to x, that the variable y is held
constant. Of course, if y is a constant, then so is y3 and, as we discovered in Chapter 4,
constants differentiate to zero. Hence
−f
5 2x 1 0 5 2x
−x
In the same way
−f
5 0 1 3y2 5 3y2
−y
This time x is held constant, so x2 goes to zero, and when we differentiate y3 with
respect to y, we get 3y2.
(b) To differentiate the function
f (x, y) = x 2 y
with respect to x, we differentiate in the normal way, taking x as the variable while
pretending that y is a constant. Now, when we differentiate a constant multiple of x 2,
we differentiate x2 to get 2x and then multiply by the constant. For example,
7x 2 differentiates to 7(2x) = 14x
−100x 2 differentiates to −100(2x) = −200x
and
cx 2 differentiates to c(2x) = 2cx
➜
394 Chapter 5 Partial Differentiation
for any constant c. In our case, y, plays the role of a constant, so
x 2 y differentiates to (2x)y = 2xy
Hence
fx = 2xy
Similarly, to find fy we treat y as the variable and x as a constant in the expression
f (x, y) = x 2y
Now, when we differentiate a constant multiple of y we just get the constant, so cy
differentiates to c. In our case, x 2 plays the role of c, so x 2y differentiates to x 2. Hence
fy = x 2
Practice Problem
2. Find expressions for the first-order partial derivatives for the functions
(a) f (x, y) = 5x 4 – y2 (b) f (x, y) = x 2y 3 – 10x
In general, when we differentiate a function of two variables, the thing we end up with is
itself a function of two variables. This suggests the possibility of differentiating a second
time. In fact there are four second-order partial derivatives. We write
−2z
2
−x
or
−2f
−x2
or
f xx
for the function obtained by differentiating twice with respect to x,
−2z
−y2
or
−2f
−y2
or
f yy
for the function obtained by differentiating twice with respect to y,
−2z
−y−x
or
−2f
−y−x
or
f yx
for the function obtained by differentiating first with respect to x and then with respect to y,
and
−2z
−x−y
or
−2f
−x−y
or
f xy
for the function obtained by differentiating first with respect to y and then with respect to x.
Section 5.1 Functions of several variables 395
Example
Find expressions for the second-order partial derivatives fxx, fyy, fyx and fxy for the
functions
(a) f (x, y) = x 2 + y3 (b) f (x, y) = x 2 y
Solution
(a) The first-order partial derivatives of the function
f (x, y) = x 2 + y3
have already been found and are given by
fx = 2x,
fy = 3y 2
To find fxx we differentiate fx with respect to x to get
fxx = 2
To find fyy we differentiate fy with respect to y to get
fyy = 6y
To find fyx we differentiate fx with respect to y to get
fyx = 0
Note how fyx is obtained. Starting with the original function
f (x, y) = x 2 + y 3
we first differentiate with respect to x to get 2x, and when we differentiate this with
respect to y we keep x constant, so it goes to zero. Finally, to find fxy we differentiate fy
with respect to x to get
fxy = 0
Note how fxy is obtained. Starting with the original function
f (x, y) = x 2 + y 3
we first differentiate with respect to y to get 3y 2, and when we differentiate this with
respect to x we keep y constant, so it goes to zero.
(b) The first-order partial derivatives of the function
f (x, y) = x 2 y
have already been found and are given by
fx = 2xy,
fy = x 2
Hence
fxx = 2y, fyy = 0
fyx = 2x,
fxy = 2x
396 Chapter 5 Partial Differentiation
Practice Problem
3. Find expressions for the second-order partial derivatives of the functions
(a) f (x, y) = 5x4 – y2 (b) f (x, y) = x 2 y 3 – 10x
[Hint: you might find your answer to Practice Problem 2 useful.]
Looking back at the expressions obtained in the previous example and Practice Problem 3,
notice that in all cases
−2 f
−2 f
5
−y−x
−x−y
fyx 5 fxy
It can be shown that this result (known as Young’s theorem) holds for all functions that
arise in economics. It is immaterial in which order the partial differentiation is performed.
Differentiating with respect to x then y gives the same expression as differentiating with
respect to y then x. (In fact, there are some weird mathematical functions for which this result
is not true, although they need not concern us.)
Although we have concentrated exclusively on functions of two variables, it should be
obvious how to work out partial derivatives of functions of more than two variables. For the
general function
y = f (x1, x2, . . . , xn)
there are n first-order partial derivatives, written as
−f
−xi
or
fi
(i 5 1, 2, . . . , n)
which are found by differentiating with respect to one variable at a time, keeping the remaining
n − 1 variables fixed. The second-order partial derivatives are determined in an analogous way.
For the function
f(x1, x2, x3) = x31 + x1x32 + 5x24
we could differentiate first with respect to x1 to get
f1 5
−f
5 3x12 1 x32
−x 1
and if we further differentiate this with respect to x3, we get
f 31 5
−2f
5 2x 3
−x 3−x 1
In fact, as we have just noted for functions of two variables, we get the same answer if we
differentiate in reverse order. You might like to check this for yourself.
Practice Problem
4. Find expressions for the partial derivatives f1, f11 and f21 in the case when
f (x1, x2, x3) = x1x2 + x 15 − x 22 x3
Section 5.1 Functions of several variables 397
We have seen how to work out partial derivatives but have yet to give any meaning to
them. To provide an interpretation of a partial derivative, let us recall the corresponding
situation for functions of one variable of the form
y = f(x)
The derivative, dy/dx, gives the rate of change of y with respect to x. In other words, if x
changes by a small amount Δx, then the corresponding change in y satisfies
Dy ù
dy
3 Dx
dx
Moreover, the accuracy of the approximation improves as Δx becomes smaller and smaller.
Advice
You might like to remind yourself of the reasoning behind this approximation, which was explained
graphically in Section 4.3.1.
Given the way in which a partial derivative is found, we can deduce that for a function of
two variables
z = f (x, y)
if x changes by a small amount Δx and y is held fixed, then the corresponding change in
z satisfies
Dz ù
−z
3 Dx
−x
Similarly, if y changes by Δy and x is fixed, then z changes by
Dz ù
−z
3 Dy
−y
In practice, of course, x and y may both change simultaneously. If this is the case, then the net
change in z will be the sum of the individual changes brought about by changes in x and y
separately, so that
Dz ù
−z
−z
Dx 1
Dy
−x
−y
This is referred to as the small increments formula. Although this is only an approximation, it
can be shown that for most functions the corresponding error tends to zero as Δx and Δy
both tend to zero. For this reason the formula is sometimes quoted with an equality sign and
written as
dz 5
−z
−z
dx 1 dy
−x
−y
where the symbols dx, dy and dz are called differentials and represent limiting values of Δx,
Δy and Δz, respectively.
398 Chapter 5 Partial Differentiation
Example
If
z = x 3 y − y 3x
evaluate
−z
−z
and
−x
−y
at the point (1, 3). Hence estimate the change in z when x increases from 1 to 1.1 and y
decreases from 3 to 2.8 simultaneously.
Solution
If
z = x 3y − y3x
then
∂z/∂x = 3x 2y − y3 and ∂z/∂y = x3 − 3y2x
so at the point (1, 3)
−z
5 3(1)2(3) 2 33 5 218
−x
−z
5 13 2 3(3)2(1) 5 226
−y
Now, since x increases from 1 to 1.1, the change in x is
Δx = 0.1
positive numbers
denote increases
and, since y decreases from 3 to 2.8, the change in y is
Δy = −0.2
negative numbers
denote decreases
The small increments formula states that
Dz ù
−z
−z
Dx 1 Dy
−x
−y
The change in z is therefore
Δz ≅ (−18)(0.1) + (−26)(−0.2) = 3.4
so z increases by approximately 3.4.
Section 5.1 Functions of several variables 399
Practice Problem
5. If
z = x y − 5x + 2y
evaluate
−z
−z
and
−x
−y
at the point (2, 6).
(a) Use the small increments formula to estimate the change in z as x decreases from 2
to 1.9 and y increases from 6 to 6.1.
(b) Confirm your estimate of part (a) by evaluating z at (2, 6) and (1.9, 6.1).
One important application of the small increments formula is to implicit differentiation.
We hope by now that you are entirely happy differentiating functions of one variable such as
y = x 3 + 2x 2 + 5
dy
5 3x2 1 4x
dx
Suppose, however, that you are asked to find dy/dx given the equation
y3 + 2xy2 − x = 5
This is much more difficult. The reason for the difference is that in the first case, y is given
explicitly in terms of x, whereas in the second case, the functional dependence of y on x is
only given implicitly. You would need to somehow rearrange this equation and to write y in
terms of x before you could differentiate it. Unfortunately, this is impossible because of the
presence of the y 3 term. The trick here is to regard the expression on the left-hand side of the
equation as a function of the two variables x and y, so that
f (x, y) = y3 + 2xy2 − x
or, equivalently,
z = y3 + 2xy2 − x
The equation
y3 + 2xy2 − x = 5
then reads
z=5
In general, the differential form of the small increments formula states that
dz 5
−z
−z
dx 1
dy
−x
−y
In our particular case, z takes the constant value of 5, so it does not change. Hence dz = 0
and the formula reduces to
05
−z
−z
dx 1
dy
−x
−y
400 Chapter 5 Partial Differentiation
which rearranges as
−z
−z
dy 5 2 dx
−y
−x
that is,
2−z/−x
dy
5
dx
−z/−y
This formula can be used to find dy/dx given any implicit function
f (x, y) = constant
that is,
if
f (x, y) 5 constant
then
dy
f
52 x
fy
dx
The technique of finding dy/dx from −fx /fy is called implicit differentiation and can be used
whenever it is difficult or impossible to obtain an explicit representation for y in terms of x.
For the function
f (x, y) = y3 + 2xy2 − x
we have
fx = 2y2 − 1 and fy = 3y2 + 4xy
so that
2y2 2 1
dy
f
22y2 1 1
52 x 52
5
3y2 1 4xy
3y2 1 4xy
dx
fy
Practice Problem
6. Use implicit differentiation to find expressions for dy/dx given that
(a) xy – y3 + y = 0 (b) y 5 – xy2 = 10
Key Terms
Dependent variable A variable whose value is determined by that taken by the independent variables; in z = f(x, y), the dependent variable is z.
−z
Differentials Limiting values of incremental changes. In the limit, the approximation Dz >
3 Dx
−
x
−z
3 dx where dz and dx are the differentials.
becomes dz 5
−x
Function of two variables A rule which assigns to each pair of incoming numbers, x and y, a
uniquely defined outgoing number, z.
Section 5.1 Functions of several variables 401
Implicit differentiation The process of obtaining dy/dx where the function is not given explicitly
as an expression for y in terms of x.
Independent variable Variables whose values determine that of the dependent variable; in
z = f(x, y), the independent variables are x and y.
Partial derivative The derivative of a function of two or more variables with respect to one of
these variables, the others being regarded as constant.
Second-order partial derivative The partial derivative of a first-order partial derivative. For
example, fxy is the second-order partial derivative when f is differentiated first with respect to y and
then with respect to x.
Small increments formula
The result Dz >
−z
−z
Dx 1 Dy
−x
−y
Exercise 5.1
1. If
f (x, y) = 3x 2y3
evaluate f (2, 3), f (5, 1) and f(0, 7).
2. If
f (x, y) = 2x2 + xy
write down an expression for
(a) f (a, a)
(b) f (b, −b)
3. If
f (x, y) = xy2 + 4x3
show that
f (2x, 2y) = 8f (x, y)
−z
−z
and , for
−x
−y
(d) z = x 6y 2 + 5y 3
4. Write down expressions for the first-order partial derivatives,
(a) z = x 2 + 4y5
(b) z = 3x 3 − 2ey
(c) z = xy + 6y
5. If
f (x, y) = x 4y5 − x 2 + y2
write down expressions for the first-order partial derivatives, fx and fy. Hence evaluate
fx(1, 0) and fy(1, 1).
6. Use the small increments formula to estimate the change in
z = x 2y 4 − x 6 + 4y
when
(a) x increases from 1 to 1.1 and y remains fixed at 0;
(b) x remains fixed at 1 and y decreases from 0 to −0.5;
(c) x increases from 1 to 1.1 and y decreases from 0 to −0.5.
➜
402 Chapter 5 Partial Differentiation
7. (a) If
f (x, y) = y − x 3 + 2x
write down expressions for fx and fy . Hence use implicit differentiation to find
dy/dx given that
y − x 3 + 2x = 1
(b) Confirm your answer to part (a) by rearranging the equation
y − x 3 + 2x = 1
to give y explicitly in terms of x and using ordinary differentiation.
8. Find the first-order partial derivatives,
(a) z = u + v2 − 5w3 + 2uv
−z −z −z
, , , for each of the following functions:
−u −v −w
(b) z = 6u1/2v1/3w1/6
9. (a) Use the small increments formula to estimate the change in
z = x3 − 2xy
when x increases from 5 to 5.5 and y increases from 8 to 8.8.
(b) By evaluating z at (5, 8) and (5.5, 8.8), work out the exact change in z, and hence
calculate the percentage error in using the small increments formula.
Exercise 5.1*
1. If
f (x, y) = 2xy + 3x
verify that f (5, 7) ≠ f(7, 5). Find all pairs of numbers, (x, y) for which f(x, y) = f(y, x).
2. If
f (w, x, y) = 5w0.34x0.25y0.41
show that f (kw, kx, ky) = kf (w, x, y)
3. Find expressions for all first- and second-order partial derivatives of the following
functions. In each case verify that
−2z
−2z
5
−y−x
−x−y
(a) z = xy (b) z = exy (c) z = x 2 + 2x + y (d) z = 16x1/4y3/4 (e) z 5
y
x
1
x2
y
4. If
z = x 2y3 − 10xy + y2
evaluate ∂z/∂x and ∂z/∂y at the point (2, 3). Hence estimate the change in z as x increases
by 0.2 and y decreases by 0.1.
Section 5.1 Functions of several variables 403
5. Find the first-order derivatives,
(a) z = (6u + vw3)4
−z −z −z
, , , for each of the following functions
−u −v −w
(b) z 5 u we2vw
6. If f (x,y) = x3e−2y evaluate all first- and second-order derivatives at (e, 1).
7. Verify that x = 1, y = −1 satisfy the equation x 2 − 2y 3 = 3. Use implicit differentiation
to find the value of dy/dx at this point.
8. A function of three variables is given by
f(x1, x2, x3) 5
x1x32
1 ln(x2x3)
x2
Find all of the first- and second-order derivatives of this function and verify that
f12 = f21,
f13 = f31
and f23 = f32
9. Write down an expression for a function, f(x, y), with first-order partial derivatives,
−f
5 3xy(xy 1 2)
−x
−f
5 x2(2xy 1 3)
−y
10. Evaluate the second-order partial derivative, f23, of the function
f(x1, x2x3) 5
x3x23
1 x2ex3
x1
at the point, (3, 2, 0).
11. Find the value of
x 2y 2
dy
at the point (−2, 1) for the function which is defined implicitly by
dx
x
56
y
12. Use implicit differentiation to find expressions, in terms of x and y, for
the following:
(a) x3y + 4xy2 = 6
(b) 12x1/3y1/4 + x = 8
(c) yexy = 10
(d)
x2 1 y2
55
x1y
dy
for each of
dx
Section 5.2
Partial elasticity and marginal
functions
Objectives
At the end of this section you should be able to:
●
Calculate partial elasticities.
●
Calculate marginal utilities.
●
Calculate the marginal rate of commodity substitution along an indifference curve.
●
Calculate marginal products.
●
Calculate the marginal rate of technical substitution along an isoquant.
●
State Euler’s theorem for homogeneous production functions.
The first section of this chapter described the technique of partial differentiation. Hopefully,
you have discovered that partial differentiation is no more difficult than ordinary differenti­
ation. The only difference is that for functions of several variables you have to be clear at the
outset which letter in a mathematical expression is to be the variable, and to bear in mind that
all remaining letters are then just constants in disguise! Once you have done this, the actual
differentiation itself obeys the usual rules. In Sections 4.3 and 4.5 we considered various
microeconomic applications. Given the intimate relationship between ordinary and partial
differentiation, you should not be too surprised to learn that we can extend these applications
to functions of several variables. We concentrate on three main areas:
●
elasticity of demand;
●
utility;
●
production.
We consider each of these in turn.
5.2.1 Elasticity of demand
Suppose that the demand, Q, for a certain good depends on its price, P, the price of an alternative good, PA, and the income of consumers, Y, so that
Q = f (P, PA, Y )
for some demand function, f.
Of particular interest is the responsiveness of demand to changes in any one of these three
variables. This can be measured quantitatively using elasticity. The (own) price elasticity of
demand is defined to be
Section 5.2 Partial elasticity and marginal functions 405
EP 5
percentage change in Q
percentage change in P
with PA and Y held constant. This definition is identical to the one given in Section 4.5, so
following the same mathematical argument presented there, we deduce that
EP 5
P
−Q
3
Q
−P
The partial derivative notation is used here because Q is now a function of several variables,
and PA and Y are held constant.
In an analogous way we can measure the responsiveness of demand to changes in the price
of the alternative good. The cross-price elasticity of demand is defined to be
EPA 5
percentage change in Q
percentage change in PA
with P and Y held constant. Again, the usual mathematical argument shows that
EPA 5
PA
−Q
3
Q
−PA
The sign of E P could turn out to be positive or negative depending on the nature of the alternative good. If the alternative good is substitutable, then Q increases as PA rises, because
consumers buy more of the given good as it becomes relatively less expensive. Consequently,
A
−Q
.0
−PA
and so E P > 0. If the alternative good is complementary, then Q decreases as PA rises, because
the bundle of goods as a whole becomes more expensive. Consequently,
A
−Q
,0
−PA
and so E P < 0.
Finally, the income elasticity of demand is defined to be
A
EY 5
percentage change in Q
percentage change in Y
and can be found from
EY 5
−Q
Y
3
Q
−Y
Again, E Y can be positive or negative. If a good is inferior, then demand falls as income rises
and EY is negative. Canned vegetables, a supermarket’s own-brand white bread and bus
transportation are examples of inferior goods. If a good is normal, then demand rises as
income rises and EY is positive. Sometimes the value of EY of a normal good might even
exceed 1. These goods are called superior. For these goods the percentage rise in consumption
is greater than the percentage increase in income. If income elasticity of demand is 1.25, a rise
of 40% in income leads to a 50% increase in consumption. Examples of superior goods
include sports cars, caviar and quality wine.
406 Chapter 5 Partial Differentiation
Example
Given the demand function
Q = 100 − 2P + PA + 0.1Y
where P = 10, PA = 12 and Y = 1000, find the
(a) price elasticity of demand;
(b) cross-price elasticity of demand;
(c) income elasticity of demand.
Is the alternative good substitutable or complementary?
Solution
We begin by calculating the value of Q when P = 10, PA = 12 and Y = 1000. The demand
function gives
Q = 100 − 2(10) + 12 + 0.1(1000) = 192
(a) To find the price elasticity of demand we partially differentiate
Q = 100 − 2P + PA + 0.1Y
with respect to P to get
−Q
5 22
−P
Hence
10
P
−Q
3 (22) 5 20.10
3
5
192
Q
−P
EP 5
(b) To find the cross-price elasticity of demand we partially differentiate
Q = 100 − 2P + PA + 0.1Y
with respect to PA to get
−Q
51
−PA
Hence
EP 5
A
12
PA
−Q
3 1 5 0.06
3
5
Q
−PA 192
The fact that this is positive shows that the two goods are substitutable.
(c) To find the income elasticity of demand we partially differentiate
Q = 100 − 2P + PA + 0.1Y
with respect to Y to get
−Q
5 0.1
−Y
Hence
EY 5
−Q
1000
Y
3 0.1 5 0.52
3
5
192
Q
−Y
Section 5.2 Partial elasticity and marginal functions 407
Practice Problem
1. Given the demand function
Q = 500 − 3P − 2PA + 0.01Y
where P = 20, PA = 30 and Y = 5000, find
(a) the price elasticity of demand;
(b) the cross-price elasticity of demand;
(c) the income elasticity of demand.
If income rises by 5%, calculate the corresponding percentage change in demand.
Would this good be classified as inferior, normal or superior?
5.2.2 Utility
So far in this text we have concentrated almost exclusively on the behaviour of producers.
In this case it is straightforward to identify the primary objective, which is to maximise profit.
We now turn our attention to consumers. Unfortunately, it is not so easy to identify the
motivation for their behaviour. One tentative suggestion is that consumers try to maximise
earned income. However, if this were the case, then individuals would try to work 24 hours
a day for seven days a week, which is not so. In practice, people like to allocate a reasonable
proportion of time to leisure activities.
Consumers are faced with a choice of how many hours each week to spend working and
how many to devote to leisure. In the same way, a consumer needs to decide how many items
of various goods to buy and has a preference between the options available. To analyse the
behaviour of consumers quantitatively, we associate with each set of options a number, U,
called utility, which indicates the level of satisfaction. Suppose that there are two goods, G1
and G2, and that the consumer buys x1 items of G1 and x2 items of G2. The variable U is then
a function of x1 and x2, which we write as
U = U(x1, x2)
If
U(3, 7) = 20 and U(4, 5) = 25
for example, then the consumer derives greater satisfaction from buying four items of G1 and
five items of G2 than from buying three items of G1 and seven items of G2.
Utility is a function of two variables, so we can work out two first-order partial derivatives,
−U
−x1
and
−U
−x2
The derivative
−U
−xi
gives the rate of change of U with respect to xi and is called the marginal utility of xi . If xi
changes by a small amount Δxi and the other variable is held fixed, then the change in U satisfies
DU >
−U
Dx
−xi i
408 Chapter 5 Partial Differentiation
If x1 and x2 both change, then the net change in U can be found from the small increments
formula
DU >
−U
−U
Dx
Dx 1
−x1 1 −x2 2
Example
Given the utility function
U = x 11/4 x 3/4
2
determine the value of the marginal utilities
−U
−x1
and
−U
−x2
when x1 = 100 and x2 = 200. Hence estimate the change in utility when x1 decreases from
100 to 99 and x2 increases from 200 to 201.
Solution
If
U = x11/4x 3/4
2
then
−U
1
x 3/4
5 x23/4
−x1
4 1 2
and
−U
3
x21/4
5 x1/4
−x2
4 1 2
so when x1 = 100 and x2 = 200
−U
5
−x1
−U
5
−x2
1
4
(100)23/4(200)3/4 5 0.42
4
(100)1/4(200)21/4 5 0.63
3
Now, x1 decreases by 1 unit, so
Δx1 = −1
and x2 increases by 1 unit, so
Δx2 = 1
The small increments formula states that
DU >
−U
−U
Dx 1
Dx
−x1 1
−x2 2
The change in utility is therefore
ΔU ≅ (0.42)(−1) + (0.63)(1) = 0.21
Section 5.2 Partial elasticity and marginal functions 409
Note that for the particular utility function
3/4
U = x 1/4
1 x2
given in the previous example, the second-order derivatives
−2U
−2U
23 27/4 3/4
23 1/4 25/4
x1 x2 and
x x
5
2 5
−x1
16
−x22
16 1 2
are both negative. Now ∂2U/∂x 21 is the partial derivative of marginal utility ∂U/∂x1 with respect
to x1. The fact that this is negative means that marginal utility of x1 decreases as x1 rises. In
other words, as the consumption of good G1 increases, each additional item of G1 bought
confers less utility than the previous item. A similar property holds for G2. This is known as
the law of diminishing marginal utility.
Advice
You might like to compare this with the law of diminishing marginal productivity discussed in
Section 4.3.2.
Practice Problem
2. An individual’s utility function is given by
U = 1000x1 + 450x2 + 5x1x2 − 2x 12 − x 22
where x1 is the amount of leisure measured in hours per week, and x2 is earned income
measured in dollars per week.
Determine the value of the marginal utilities
−U
−x1
and
−U
−x2
when x1 = 138 and x2 = 500.
Hence estimate the change in U if the individual works for an extra hour, which
increases earned income by $15 per week.
Does the law of diminishing marginal utility hold for this function?
It was pointed out in Section 5.1 that functions of two variables could be represented by
surfaces in three dimensions. This is all very well in theory, but in practice the task of sketching
such a surface by hand is virtually impossible. This difficulty has been faced by geographers
for years, and the way they circumvent the problem is to produce a two-dimensional contour
map. A contour is a curve joining all points at the same height above sea level. Exactly the
same device can be used for utility functions. Rather than attempt to sketch the surface, we
draw an indifference map. This consists of indifference curves joining points (x1, x2) which give
the same value of utility. Mathematically, an indifference curve is defined by an equation
U(x1, x2) = U0
for some fixed value of U0. A typical indifference map is sketched in Figure 5.4.
Points A and B both lie on the lower indifference curve, U0 = 20. Point A corresponds to the
case when the consumer buys a1 units of G1 and a2 units of G2. Likewise, point B corresponds
410 Chapter 5 Partial Differentiation
x2
a2
A
D
C
U0 5 40
B
b2
U0 5 30
U0 5 20
a1
b1
x1
Figure 5.4
to the case when the consumer buys b1 units of G1 and b2 units of G2. Both of these combinations yield the same level of satisfaction and the consumer is indifferent to choosing between
them. In symbols we have
U(a1, a2) = 20 and U(b1, b2) = 20
Points C and D lie on indifference curves that are further away from the origin. The combinations of goods that these points represent yield higher levels of utility and so are ranked above
those of A and B.
Indifference curves are usually downward-sloping. If fewer purchases are made of G1, then
the consumer has to compensate for this by buying more of type G2 to maintain the same
level of satisfaction. Note also from Figure 5.4 that the slope of an indifference curve varies
along its length, taking large negative values close to the vertical axis and becoming almost
zero as the curve approaches the horizontal axis. Again, this is to be expected for any function that obeys the law of diminishing marginal utility. A consumer who currently owns a
large number of items of G2 and relatively few of G1 is likely to value G1 more highly.
Consequently, he or she might be satisfied in sacrificing a large number of items of G2 to gain
just one or two extra items of G1. In this region the marginal utility of x1 is much greater than
that of x2, which accounts for the steepness of the curve close to the vertical axis. Similarly, as
the curve approaches the horizontal axis, the situation is reversed and the curve flattens off.
We quantify this exchange of goods by introducing the marginal rate of commodity substitution,
MRCS. This is defined to be the increase in x2 necessary to maintain a constant value of utility
when x1 decreases by 1 unit. This is illustrated in Figure 5.5.
Starting at point E, we move 1 unit to the left. The value of MRCS is then the vertical dis­
tance that we need to travel if we are to remain on the indifference curve passing through E.
Now this sort of ‘1-unit change’ definition is precisely the approach that we took in Section 4.3
when discussing marginal functions. In that section we actually defined the marginal function
to be the derived function and we showed that the ‘1-unit change’ definition gave a good
approximation to it. If we do the same here, then we can define
MRCS 5 2
dx2
dx1
The derivative, dx2/dx1, determines the slope of an indifference curve when x1 is plotted on
the horizontal axis and x2 is plotted on the vertical axis. This is negative, so we deliberately
SectiOn 5.2 PArTIAL eLASTICITY AND mArgINAL FUNCTIONS
411
x2
F
MRCS
E
indifference curve
1
x1
Figure 5.5
put a minus sign in front to make MRCS positive. This definition is useful only if we can find
the equation of an indifference curve with x2 given explicitly in terms of x1. However, we may
only know the utility function
U(x1, x2)
so that the indifference curve is determined implicitly from an equation
U(x1, x2) = U0
This is precisely the situation that we discussed at the end of Section 5.1. The formula for
implicit differentiation gives
dx2
−U/−x1
52
dx1
−U/−x2
Hence
MRCS 5 2
dx2
−U/−x1
5
dx1
−U/−x2
marginal rate of commodity substitution is the marginal
utility of x1 divided by the marginal utility of x2
example
Given the utility function
U 5 x11/2x21/2
find a general expression for MRCS in terms of x1 and x2.
Calculate the particular value of MRCS for the indifference curve that passes through
(300, 500). Hence estimate the increase in x2 required to maintain the current level of utility when x1 decreases by 3 units.
➜
412 Chapter 5 Partial Differentiation
Solution
If
U 5 x11/2x21/2
then
−U
1
x 1/2 and
5 x21/2
−x1
2 1 2
−U
1
x21/2
5 x1/2
−x2
2 1 2
Using the result
MRCS 5
−U/−x1
−U/−x2
we see that
x21/2
x21/2
1
1/2
1 x x21/2
2 1
2
1
MRCS 5
2
1
5 x21
1 x2
5
x2
x1
At the point (300, 500)
MRCS 5
rule 2 of indices;
bm 4 bn 5 bm2n
b1 5 b,
b21 5
1
b
500
5
5
300
3
Now MRCS approximates the increase in x2 required to maintain a constant level of utility when x1 decreases by 1 unit. In this example x1 decreases by 3 units, so we multiply MRCS
by 3. The approximate increase in x2 is
5
3355
3
We can check the accuracy of this approximation by evaluating U at the old point
(300, 500) and the new point (297, 505). We get
U(300, 500) = (300)1/2 (500)1/ 2 = 387.30
U(297, 505) = (297)1/2 (505)1/ 2 = 387.28
This shows that, to all intents and purposes, the two points do indeed lie on the same
indifference curve.
Practice Problem
3. Calculate the value of MRCS for the utility function given in Practice Problem 2 at the
point (138, 500). Hence estimate the increase in earned income required to maintain
the current level of utility if leisure time falls by two hours per week.
SectiOn 5.2 PArTIAL eLASTICITY AND mArgINAL FUNCTIONS
413
5.2.3 Production
Production functions were first introduced in Section 2.3. We assume that output, Q, depends
on capital, K, and labour, L, so we can write
Q = f (K, L)
Such functions can be analysed in a similar way to utility functions. The partial derivative
−Q
−K
gives the rate of change of output with respect to capital and is called the marginal product of
capital, MPK. If capital changes by a small amount ΔK, with labour held constant, then the
corresponding change in Q is given by
DQ >
−Q
DK
−K
Similarly,
−Q
−L
gives the rate of change of output with respect to labour and is called the marginal product
of labour, MPL. If labour changes by a small amount Δ L, with capital held constant, then the
corresponding change in Q is given by
DQ >
−Q
DL
−L
If K and L both change simultaneously, then the net change in Q can be found from the small
increments formula
DQ >
−Q
−Q
DK 1
DL
−K
−L
The contours of a production function are called isoquants. In Greek ‘iso’ means ‘equal’, so
the word ‘isoquant’ literally translates as ‘equal quantity’. Points on an isoquant represent
all possible combinations of inputs (K, L) which produce a constant level of output, Q0. A
typical isoquant map is sketched in Figure 5.6. Notice that we have adopted the standard
convention of plotting labour on the horizontal axis and capital on the vertical axis.
The lower curve determines the input pairs needed to output 100 units. Higher levels of
output correspond to isoquants further away from the origin. Again, the general shape of the
curves is to be expected. For example, as capital is reduced, it is necessary to increase labour
to compensate and so maintain production levels. Moreover, if capital continues to decrease,
the rate of substitution of labour for capital goes up. We quantify this exchange of inputs by
defining the marginal rate of technical substitution, MRTS, to be
2
dK
dL
so that MRTS is the positive value of the slope of an isoquant. As in the case of a utility function, the formula for implicit differentiation shows that
MRTS 5
−Q/−L
MPL
5
−Q/−K
MPK
marginal rate of technical substitution is the marginal product
of labour divided by the marginal product of capital
414 Chapter 5 Partial Differentiation
K
isoquant
Q0 5 200
Q0 5 150
Q0 5 100
L
Figure 5.6
As an example, consider the Cobb–Douglas production function given by
Q = AK aLb
where A, a and b are positive constants. Partial differentiation of
Q = AK aLb
with respect to K and L gives
MPK = aAK a−1Lb and MPL = bAK aLb−1
Hence
MPL
bAKaLb21
bK
5
5
MPK
aL
aAKa21Lb
MRTS 5
Practice Problem
4. Given the production function
Q = K 2 + 2L2
write down expressions for the marginal products
−Q
−K
and
−Q
−L
Hence show that
2L
K
−Q
−Q
(b) K
5 2Q
1L
−K
−L
(a) MRTS 5
Advice
Production functions and the concept of homogeneity were covered in Section 2.3. You might find
it useful to revise this before reading the next paragraph.
Section 5.2 Partial elasticity and marginal functions 415
Recall that a production function is described as being homogeneous of degree n if, for any
number l,
f (lK, lL) = lnf (K, L)
A production function is then said to display decreasing returns to scale, constant returns
to scale or increasing returns to scale, depending on whether n < 1, n = 1 or n > 1, respectively.
One useful result concerning homogeneous functions is known as Euler’s theorem, which
states that
K
−f
−f
5 nf(K, L)
1L
−K
−L
In fact, you have already verified this in Practice Problem 4(b) for the particular production
function
Q = K 2 + 2L2
which is easily shown to be homogeneous of degree 2. We have no intention of proving this
theorem, although you are invited to confirm its validity for general Cobb–Douglas production functions in Question 4 in Exercise 5.2* at the end of this section.
The special case n = 1 is worthy of note because the right-hand side is then simply f(K, L),
which is equal to the output, Q. Euler’s theorem for homogeneous production functions of
degree 1 states that
labour times marginal
Capital times marginal
5 total output
1
product of labour
product of capital
If each input factor is paid an amount equal to its marginal product, then each term on the
left-hand side gives the total bill for that factor. For example, if each unit of labour is paid
MPL, then the cost of L units of labour is L(MPL). Provided that the production function displays constant returns to scale, Euler’s theorem shows that the sum of the factor payments is
equal to the total output.
Key Terms
Cross-price elasticity of demand The responsiveness of demand for one good due to a change
in the price of another: (percentage change in quantity) ÷ (percentage change in the price of the
alternative good).
Euler’s theorem If each input is paid the value of its marginal product, the total cost of these
inputs is equal to total output, provided there are constant returns to scale.
Income elasticity of demand The responsiveness of demand for one good due to a change in
income: (percentage change in quantity) ÷ (percentage change in income).
Indifference curve
utility.
A curve indicating all combinations of two goods which give the same level of
Indifference map A diagram showing the graphs of a set of indifference curves. The further the
curve is from the origin, the greater the level of utility.
Isoquant
A curve indicating all combinations of two factors which give the same level of output.
Law of diminishing marginal utility The law which states that the increase in utility due to the
consumption of an additional good will eventually decline: ∂2U/∂x2i < 0 for sufficiently large xi .
Marginal product of capital
MPK = ∂Q/∂K.
The additional output produced by a 1 unit increase in capital:
➜
416 Chapter 5 Partial Differentiation
Marginal product of labour
MPL = ∂Q/∂L.
The additional output produced by a 1-unit increase in labour:
Marginal rate of commodity substitution (MRCS) The amount by which one input needs to
increase to maintain a constant value of utility when the other input decreases by 1 unit:
MRCS = ∂U/∂x1 ÷ ∂U/∂x2.
Marginal rate of technical substitution (MRTS) The amount by which capital needs to rise to
maintain a constant level of output when labour decreases by 1 unit: MRTS = MPL /MPK.
Marginal utility
The extra satisfaction gained by consuming 1 extra unit of a good: ∂U/∂xi .
Price elasticity of demand A measure of the responsiveness of the change in demand due to a
change in price: (percentage change in demand) ÷ (percentage change in price).
Superior good A normal good for which the percentage rise in consumption exceeds the percentage increase in income.
Utility
The satisfaction gained from the consumption of a good.
Exercise 5.2
1. Given the demand function
Q = 1000 − 5P − PA2 + 0.005Y 3
where P = 15, PA = 20 and Y = 100, find the income elasticity of demand and explain
why this is a superior good.
Give your answer correct to two decimal places.
2. Given the demand function
Q = 200 − 2P − PA + 0.1Y 2
where P = 10, PA = 15 and Y = 100, find
(a) the price elasticity of demand;
(b) the cross-price elasticity of demand;
(c) the income elasticity of demand.
Estimate the percentage change in demand if PA rises by 3%. Is the alternative good
substitutable or complementary?
3. A utility function is given by U = 2x2 + y2.
(a) State the equation of the indifference curve which passes through (4, 2).
(b) Calculate the marginal utilities at (4, 2) and hence work out the gradient of the
curve at this point.
4. The satisfaction gained by consuming x units of good 1 and y units of good 2 is meas-
ured by the utility function
U = 2x2 + 5y3
Currently an individual consumes 20 units of good 1 and 8 units of good 2.
(a) Find the marginal utility of good 1 and hence estimate the increase in satisfaction
gained from consuming one more unit of good 1.
(b) Find the marginal utility of good 2 and hence estimate the increase in satisfaction
gained from consuming one more unit of good 2.
Section 5.2 Partial elasticity and marginal functions 417
5. Given the demand function
Q5
PAY
P2
find the income elasticity of demand.
6. Given the utility function
U = x 11/2x 1/3
2
determine the value of the marginal utilities
−U
−x1
and
−U
−x2
at the point (25, 8). Hence
(a) estimate the change in utility when x1 and x2 both increase by 1 unit;
(b) find the marginal rate of commodity substitution at this point.
7. Evaluate MPK and MPL for the production function
Q 5 2LK 1 L
given that the current levels of K and L are 7 and 4, respectively. Hence
(a) write down the value of MRTS;
(b) estimate the increase in capital needed to maintain the current level of output given
a 1-unit decrease in labour.
8. If Q = 2K 3 + 3L2K, show that K(MPK ) + L(MPL) = 3Q.
9. The demand functions for two commodities, A and B, are given by
QA = AP −0.5Y 0.5 and QB = BP −1.5Y 1.5
where A and B are positive constants
(a) Find the price elasticity of demand for each good and hence comment on the rela-
tive sensitivity of demand due to changes in price.
(b) Find the income elasticity of demand for each good. Which good is normal and
which is superior? Give a reason for your answer.
10. A firm’s production function is given by Q = 18K1/6L5/6.
(a) Show that this function displays constant returns to scale.
(b) Find expressions for the marginal products of capital and labour.
(c) State what happens to the marginal product of labour when
(i) labour increases with capital held constant;
(ii) capital increases with labour held constant.
418 Chapter 5 Partial Differentiation
Exercise 5.2*
1. The demand function of a good is given by
Q = 500 − 4P + 0.02Y
Price and income are known to be P = 20 and Y = 14 000, respectively.
(a) Find the income elasticity of demand.
(b) Estimate the percentage change in demand when income rises by 8%, and comment
on the growth potential of this good in an expanding economy.
2. Find the value of the marginal rate of technical substitution for the production function,
Q = 300K 2/3L1/2
when K = 40, L = 60.
3. A utility function is given by
U = x 12/3x 21/2
Find the value of x2 if the points (64, 256) and (512, x 2) lie on the same indifference curve.
4. Show that the Cobb–Douglas production function Q = AKaLb is homogeneous of degree
a + b and that
(a) K
−Q
−Q
5 (a 1 b)Q
1L
−L
−K
(b) K 2
−2Q
−2Q
−2Q
2
1
2KL
1
L
5 (a 1 b)(a 1 b 2 1)Q
−K 2
−K−L
−L2
5. An individual’s utility function is given by
U = Ax10.7x20.5
where x1 and x2 denote the number of units consumed of goods 1 and 2.
(a) Show that the marginal utility of x1 is positive and give an interpretation of this result.
−2 U
(b) Show that the second-order derivative
is positive and give an interpretation
−x1−x2
of this result.
−2U
(c) Show that the second-order derivative −x2 is negative and give an interpretation of
1
this result.
6. If a firm’s production function is given by
Q = 5L + 7K
sketch the isoquant corresponding to an output level, Q = 700. Use your graph to find
the value of MRTS and confirm this using partial differentiation.
7. A firm’s production function is given by
Q 5 10 (KL) 1 3L
with K = 90 and L = 40.
(a) Find the values of the marginal products, MPK and MPL.
(b) Use the results of part (a) to estimate the overall effect on Q when K increases by
3 units and L decreases by 2 units.
(c) State the value of the marginal rate of technical substitution and give an interpreta-
tion of this value.
Section 5.2 Partial elasticity and marginal functions 419
8. A firm’s production function is
Q = A[bK a + (1 − b)La]1/a
(a) Show that the marginal rate of technical substitution is given by
12b K
MRTS 5
b
L
12a
(b) Show that the marginal products satisfy the relation
K
−Q
−Q
5Q
1L
−K
−L
9. The demand function of a good is given by
Q = a − bP − cPA + dY
where P is the price of the good, PA is the price of an alternative good, Y is income and
the coefficients a, b, c and d are all positive. It is known that P = 50, PA = 30, Y = 1000
and Q = 5000.
(a) Is the alternative good substitutable or complementary? Give a reason for your
answer.
(b) Find expressions, in terms of b, c and d, for the
(i) price elasticity of demand;
(ii) cross-price elasticity of demand;
(iii) income elasticity of demand.
(c) The cross-price elasticity is −0.012. The income elasticity is four times the magni-
tude of the price elasticity. When income increases by 10%, the demand increases
by 2%. Determine the values of a, b, c and d.
10. The demand function of a good is given by
Q = kP−aPAb Y c
where P is the own price, PA is the price of an alternative good and Y is income. The
letters k, a, b and c are positive constants.
(a) Explain why the alternative good is substitutable.
(b) Show that own price elasticity, cross-price elasticity and income elasticity are −a, b
and c, respectively.
(c) Explain briefly why the proportion of income spent on this good is given by PQ/Y
and deduce that for this demand function, this proportion is
kP1−aPAb Y c−1
Hence show that when c > 1, the proportion of income spent on this good rises with
income.
11. Find the marginal rate of commodity substitution for each of these utility functions
and hence state whether the indifference map consists of convex curves, concave curves
or straight lines.
(a) U = (2x1 + 3x2)3 (b) U = 5x13x2 (c) U 5 2 x1 1 6 x2
Section 5.3
Comparative statics
Objectives
At the end of this section you should be able to:
●
Use structural equations to derive the reduced form of macroeconomic models.
●
Calculate national income multipliers.
●
Use multipliers to give a qualitative description of economic models.
●
Use multipliers to give a quantitative description of economic models.
●
Calculate multipliers for the linear one-commodity market model.
Advice
The content of this section is quite difficult since it depends on ideas covered earlier in this
text. You might find it helpful to read quickly through Section 1.7 now before tackling the new
material.
The simplest macroeconomic model, discussed in Section 1.7, assumes that there are two
sectors, households and firms, and that household consumption, C, is modelled by a linear
relationship of the form
C = aY + b
(1)
In this equation Y denotes national income and a and b are parameters. The parameter a is
the marginal propensity to consume and lies in the range 0 < a < 1. The parameter b is the
autonomous consumption and satisfies b > 0. In equilibrium
Y=C+I
(2)
where I denotes investment, which is assumed to be given by
I = I*
(3)
for some constant I*. Equations (1), (2) and (3) describe the structure of the model and as
such are called structural equations. Substituting equations (1) and (3) into equation (2) gives
Y = aY + b + I*
Y − aY = b + I *
(1 − a)Y = b + I*
Y5
b 1 I*
12a
(subtract aY from both sides)
(take out a common factor of Y)
(divide both sides by 1 2 a)
Section 5.3 Comparative statics 421
This is known as the reduced form because it compresses the model into a single equation
in which the endogenous variable, Y, is expressed in terms of the exogenous variable, I*, and
parameters, a and b. The process of analysing the equilibrium level of income in this way is
referred to as statics because it assumes that the equilibrium state is attained instantaneously.
The branch of mathematical economics which investigates time dependence is known as
dynamics and is considered in Chapter 9.
We should like to do rather more than just to calculate the equilibrium values here. In
particular, we are interested in the effect on the endogenous variables in a model brought
about by changes in the exogenous variables and parameters. This is known as comparative
statics, since we seek to compare the effects obtained by varying each variable and parameter
in turn. The actual mechanism for change will be ignored, and it will be assumed that the
system returns to equilibrium instantaneously. The equation
Y5
b 1 I*
12a
shows that Y is a function of three variables, a, b and I*, so we can write down three partial
derivatives
−Y −Y −Y
, ,
−a −b −I*
The only hard one to work out is the first, and this is found using the chain rule by writing
Y = (b + I*)(1 − a)−1
which gives
−Y
b 1 I*
5 (b 1 I*)(21)(1 2 a)22 (21) 5
−a
(1 2 a)2
To interpret this derivative let us suppose that the marginal propensity to consume, a,
changes by Δa with b and I* held constant. The corresponding change in Y is given by
DY 5
−Y
Da
−a
Strictly speaking, the ‘=’ sign should really be ‘≅’. However, as we have seen in the previous
two sections, provided that Δa is small, the approximation is reasonably accurate. In any case,
we could argue that the model itself is only a first approximation to what is really happening
in the economy and so any further small inaccuracies that are introduced are unlikely to have
any significant effect on our conclusions. The above equation shows that the change in
national income is found by multiplying the change in the marginal propensity to consume
by the partial derivative ∂Y/∂a. For this reason, the partial derivative is called the marginal
propensity to consume multiplier for Y. In the same way, ∂Y/∂b and ∂Y/∂I* are called the
autonomous consumption multiplier and the investment multiplier, respectively.
Multipliers enable us to explain the behaviour of the model both qualitatively and quantitatively. The qualitative behaviour can be described simply by inspecting the multipliers
as they stand, before any numerical values are assigned to the variables and parameters. It is
usually possible to state whether the multipliers are positive or negative and hence whether
an increase in an exogenous variable or parameter leads to an increase or decrease in the
corresponding endogenous variable. In the present model it is apparent that the marginal
propensity to consume multiplier for Y is positive because it is known that b and I* are both
positive, and the denominator (1 − a)2 is clearly positive. Therefore, national income rises
whenever a rises.
422 Chapter 5 Partial Differentiation
Once the exogenous variables and parameters have been assigned specific numerical
values, the behaviour of the model can be explained quantitatively. For example, if b = 10,
I* = 30 and a = 0.5, then the marginal propensity to consume multiplier is
b 1 I*
10 1 30
5
5 160
(1 2 0.5)2
(1 2 a)2
This means that when the marginal propensity to consume rises by, say, 0.02 units, the
change in national income is
160 × 0.02 = 3.2
Of course, if a, b and I* change by amounts Δa, Δb and ΔI* simultaneously, then the small
increments formula shows that the change in Y can be found from
DY 5
−Y
−Y
−Y
Da 1
Db 1
DI*
−a
−b
−I*
The investment multiplier in this model can be found by first writing
Y5
b
I*
1
12a 12a
so that
−Y
1
5
−I* 1 2 a
which is positive because a < 1. To analyse this multiplier quantitatively let us suppose that
the marginal propensity to consume, a = 0.6, and consider what happens to national income
when investment rises by 4 units. In this case the investment multiplier is
1
1
1
5
5
5 2.5
1 2 a 1 2 0.6 0.4
so if investment were to rise by 4 units, the change in national income would be 2.5 × 4 = 10.
Practice Problem
1. By substituting
Y5
b 1 I*
12a
into
C = aY + b
write down the reduced equation for C in terms of a, b and I*. Hence show that the
investment multiplier for C is
a
12a
Deduce that an increase in investment always leads to an increase in consumption.
Calculate the change in consumption when investment rises by 2 units if the marginal
propensity to consume is 1/2 .
Section 5.3 Comparative statics 423
The following example is more difficult because it involves three sectors: households,
firms and government. However, the basic strategy for analysing the model is the same. We
first obtain the reduced form, which is differentiated to determine the relevant multipliers.
These can then be used to discuss the behaviour of national income both qualitatively and
quantitatively.
Example
Consider the three-sector model
Y = C + I + G
(1)
C = aYd + b
(2)
(0 < a < 1, b > 0)
Yd = Y − T
(3)
T = tY + T*
(0 < t < 1, T* > 0)
(4)
I = I*
(I* > 0)
(5)
G = G*
(G* > 0)
(6)
where G denotes government expenditure and T denotes taxation.
(a) Show that
Y5
2aT* 1 b 1 I* 1 G*
1 2 a 1 at
(b) Write down the government expenditure multiplier and autonomous taxation multi-
plier. Deduce the direction of change in Y due to increases in G* and T*.
(c) If it is government policy to finance any increase in expenditure, ΔG*, by an increase
in autonomous taxation, ΔT*, so that
ΔG* = ΔT*
show that national income rises by an amount that is less than the rise in expenditure.
(d) If a = 0.7, b = 50, T* = 200, t = 0.2, I* = 100 and G* = 300, calculate the equilibrium
level of national income, Y, and the change in Y due to a 10-unit increase in government expenditure.
Solution
(a) We need to ‘solve’ equations (1)–(6) for Y. An obvious first move is to substitute
equations (2), (5) and (6) into equation (1) to get
Y = aYd + b + I* + G*
(7)
Now, from equations (3) and (4),
Yd = Y − T
= Y − (tY + T*)
= Y − tY − T *
so this can be put into equation (7) to get
Y = a(Y − tY − T *) + b + I* + G*
= aY − atY − aT * + b + I * + G*
➜
424 Chapter 5 Partial Differentiation
Collecting terms in Y on the left-hand side gives
(1 − a + at)Y = −aT * + b + I* + G*
which produces the desired equation
Y5
2aT* 1 b 1 I* 1 G*
1 2 a 1 at
(b) The government expenditure multiplier is
−Y
1
5
−G*
1 2 a 1 at
and the autonomous taxation multiplier is
2a
−Y
5
−T*
1 2 a 1 at
We are given that a < 1, so 1 − a > 0. Also, we know that a and t are both positive, so
their product, at, must be positive. The expression (1 – a) + at is therefore positive,
being the sum of two positive terms. The government expenditure multiplier is
therefore positive, which shows that any increase in G* leads to an increase in Y.
The autonomous taxation multiplier is negative because its numerator is negative
and its denominator is positive. This shows that any increase in T * leads to a decrease
in Y.
(c) Government policy is to finance a rise in expenditure out of autonomous taxation, so
that
ΔG* = ΔT*
From the small increments formula
DY 5
−Y
−Y
DG* 1
DT *
−G*
−T *
we deduce that
DY 5
−Y
−Y
1
DG* 5
−T*
−G*
1
a
1
DG* 5
1 2 a 1 at
1 2 a 1 at
12a
DG*
1 2 a 1 at
The multiplier
12a
1 2 a 1 at
is called the balanced budget multiplier and is positive because the numerator and
denominator are both positive. An increase in government expenditure leads to an
increase in national income. However, the denominator is greater than the numerator
by an amount at, so that
12a
,1
1 2 a 1 at
and ΔY < ΔG*, showing that the rise in national income is less than the rise in expenditure.
Section 5.3 Comparative statics 425
(d) To solve this part of the problem we simply substitute the numerical values a = 0.7,
b = 50, T* = 200, t = 0.2, I* = 100 and G* = 300 into the results of parts (a) and (b).
From part (a)
Y5
20.7(200) 1 50 1 100 1 300
2aT* 1 b 1 I* 1 G*
5 704.5
5
1 2 0.7 1 0.7(0.2)
1 2 a 1 at
From part (b) the government expenditure multiplier is
1
1
5
5 2.27
1 2 a 1 at
0.44
and we are given that ΔG* = 10, so the change in national income is
2.27 × 10 = 22.7
Practice Problem
2. Consider the four-sector model
Y=C+I+G+X−M
C = aY + b
(0 < a < 1, b > 0)
I = I*
(I* > 0)
G = G*
(G* > 0)
X = X*
(X* > 0)
M = mY + M*
(0 < m < 1, M* > 0)
where X and M denote exports and imports, respectively, and m is the marginal propensity to import.
(a) Show that
Y5
b 1 I* 1 G* 1 X* 2 M*
12a1m
(b) Write down the autonomous export multiplier
−Y
−X*
and the marginal propensity to import multiplier
−Y
−m
Deduce the direction of change in Y due to increases in X* and m.
(c) If a = 0.8, b = 120, I* = 100, G* = 300, X* = 150, m = 0.1 and M* = 40, calculate
the equilibrium level of national income, Y, and the change in Y due to a 10-unit
increase in autonomous exports.
426 Chapter 5 Partial Differentiation
P
d
supply curve
equilibrium
b
demand curve
Q
Figure 5.7
So far, all of the examples of comparative statics that we have considered have been taken
from macroeconomics. The same approach can be used in microeconomics. For example, let
us analyse the equilibrium price and quantity in supply and demand theory.
Figure 5.7 illustrates the simple linear one-commodity market model described in
Section 1.5. The equilibrium values of price and quantity are determined from the point of
intersection of the supply and demand curves. The supply curve is a straight line with a
positive slope and intercept, so its equation may be written as
P = aQS + b
(a > 0, b > 0)
The demand equation is also linear but has a negative slope and a positive intercept, so its
equation may be written as
P = −cQD + d
(c > 0, d > 0)
It is apparent from Figure 5.7 that in order for these two lines to intersect in the positive
quadrant, it is necessary for the intercept on the demand curve to lie above that on the supply
curve, so we require
d>b
or equivalently
d−b>0
In equilibrium QS and QD are equal. If we let their common value be denoted by Q, then the
supply and demand equations become
P = aQ + b
P = −cQ + d
and so
aQ + b = −cQ + d
since both sides are equal to P.
To solve for Q we first collect like terms together, which gives
(a + c)Q = d − b
Section 5.3 Comparative statics 427
and then divide by the coefficient of Q to get
Q5
d2b
a1c
(Incidentally, this confirms the restriction d − b > 0. If this were not true, then Q would be
either zero or negative, which does not make economic sense.)
Equilibrium quantity is a function of the four parameters a, b, c and d, so there are four
multipliers
−Q
d2b
52
−a
(a 1 c)2
1
−Q
52
−b
a1c
−Q
d2b
52
−c
(a 1 c)2
−Q
1
5
−d
a1c
where the chain rule is used to find ∂Q/∂a and ∂Q/∂c.
We noted previously that all of the parameters are positive and that d − b > 0, so
−Q
, 0,
−a
−Q
, 0,
−b
−Q
, 0 and
−c
−Q
.0
−d
This shows that an increase in a, b or c causes a decrease in Q, whereas an increase in d causes
an increase in Q.
It is possible to confirm the signs of these multipliers graphically. For example, from the
supply equation
P = aQS + b
we see that a small increase in the value of the parameter a causes the supply curve to become
slightly steeper, as indicated by the dashed line in Figure 5.8. The effect is to shift the point
of intersection to the left, and so the equilibrium quantity decreases from Q1 to Q2 which is
consistent with a negative value of the multiplier, ∂Q/∂a.
P
d
steeper
supply
curve
same
demand
curve
b
Q2 Q1
Figure 5.8
Q
428 Chapter 5 Partial Differentiation
Given any pair of supply and demand equations, we can easily calculate the effect on the
equilibrium quantity. For example, consider the equations
P = QS + 1
P = −2QD + 5
and let us suppose that we need to calculate the change in equilibrium quantity when the
coefficient of QS increases from 1 to 1.1. In this case we have
a = 1, b = 1,
c = 2,
d=5
To find ΔQ we first evaluate the multiplier
d2b
−Q
521
52
52
5 20.44
2
−a
(a 1 c)
(1 1 2)2
and then multiply by 0.1 to get
ΔQ = (−0.44) × 0.1 = −0.044
An increase of 0.1 in the slope of the supply curve therefore produces a decrease of 0.044 in
the equilibrium quantity.
Practice Problem
3. Give a graphical confirmation of the sign of the multiplier
−Q
−d
for the linear one-commodity market model
P = aQS + b (a > 0, b > 0)
P = −cQD + d (c > 0, d > 0)
Throughout this section all of the relations in each model have been assumed to be linear.
It is possible to analyse non-linear relations in a similar way, although this is beyond the
scope of this text.
Advice
We shall return to this topic again in Chapter 7 when we use Cramer’s rule to solve the
structural equations of a linear model.
Section 5.3 Comparative statics 429
Key Terms
Autonomous consumption multiplier The number by which you multiply the change in autonomous consumption to deduce the corresponding change in, say, national income: ∂Y/∂b.
Balanced budget multiplier The number by which you multiply the change in government
expenditure to deduce the corresponding change in, say, national income, ∂Y/∂G*, assuming that
this change is financed entirely by a change in taxation.
Comparative statics Examination of the effect on equilibrium values due to changes in the
parameters of an economic model.
Dynamics
Analysis of how equilibrium values vary over time.
Investment multiplier The number by which you multiply the change in investment to deduce the
corresponding change in, say, national income: ∂Y/∂I*.
Marginal propensity to consume multiplier The number by which you multiply the change in
MPC to deduce the corresponding change in, say, national income: ∂Y/∂a.
Reduced form The final equation obtained when endogenous variables are eliminated in the
course of solving a set of structural equations in a macroeconomic model.
Statics The determination of the equilibrium values of variables in an economic model which do
not change over time.
Structural equations A collection of equations that describe the equilibrium conditions of a
macroeconomic model.
Exercise 5.3*
1. Consider the three-sector model
Y = C + I + G
(1)
C = aYd + b
(2)
(0 < a < 1, b > 0)
Yd = Y − T
(3)
T = T*
(T * > 0)
(4)
I = I*
(I* > 0)
(5)
G = G*
(G* > 0)
(6)
(a) Show that
C5
aI* 1 aG* 2 aT * 1 b
12a
(b) Write down the investment multiplier for C. Decide the direction of change in C
due to an increase in I *.
(c) If a = 0.9, b = 80, I* = 60, G* = 40 and T * = 20, calculate the equilibrium level of
consumption, C, and also the change in C due to a 2-unit change in investment.
2. The reduced form of a macroeconomic model is
Y5
b 1 I* 1 G* 2 aT *
1 2 a 2 at
where t is the marginal rate of taxation.
Find an expression for the marginal rate of taxation multiplier.
➜
430 Chapter 5 Partial Differentiation
3. Consider the four-sector macroeconomic model
Y=C+I+G+X−M
C = aYd + b
(0 < a < 1, b > 0)
Yd = Y − T
T = tY + T*
(0 < t < 1, T* > 0)
I = I*
(I* > 0)
G = G*
(G* > 0)
X = X*
(X* > 0)
M = mYd + M*
(0 < m < 1, M* > 0)
(1) Show that
Y5
b 1 (m 2 a)T * 1 I* 1 G* 1 X* 2 M*
1 2 a 1 at 1 m 2 mt
(2) (a) Write down the autonomous taxation multiplier. Deduce that an increase in
T* causes a decrease in Y on the assumption that a country’s marginal propensity to import, m, is less than its marginal propensity to consume, a.
(b) Write down the government expenditure multiplier. Deduce that an increase in
G* causes an increase in Y.
(3) Let a = 0.7, b = 150, t = 0.25, m = 0.1, T* = 100, I* = 100, G* = 500, M* = 300 and
X* = 160.
(a) Calculate the equilibrium level of national income.
(b) Calculate the change in Y due to an 11-unit increase in G*.
(c) Find the increase in autonomous taxation required to restore Y to its level calcu-
lated in part (a).
4. Show that the equilibrium price for a linear one-commodity market model
P = aQS + b
(a > 0, b > 0)
P = −cQD + d
(c > 0, d > 0)
where d – b > 0, is given by
P5
ad 1 bc
a1c
Find expressions for the multipliers
−P −P −P −P
,
,
,
−a −b −c −d
and deduce the direction of change in P due to an increase in a, b, c or d.
Section 5.3 Comparative statics 431
5. Consider the three-sector macroeconomic model
Y=C+I+G
C = aYd + b
(0 < a < 1, b > 0)
Yd = Y − T
T = T*
(T * > 0)
I = I*
(I * > 0)
G = G*
(G* > 0)
(a) Show that
Y5
1
(b 2 aT* 1 I* 1 G*)
12a
(b) Write down expressions for the government expenditure multiplier, ∂Y/∂G*, and
the taxation multiplier, ∂Y/∂T*, and deduce that if both government expenditure
and taxation increase by 1 unit, then the equilibrium value of income also rises by
1 unit, irrespective of the value of a.
(c) State the value of the balanced budget multiplier.
6. (1) For the commodity market
Y=C+I
C = aY + b
(0 < a < 1, b > 0)
I = cr + d
(c < 0, d > 0)
where r is the interest rate, show that when the commodity market is in equilibrium,
(1 − a)Y − cr = b + d
(2) For the money market
(money supply)
MS = M*S
(M*S > 0)
(total demand for money)
MD = k1Y + k2r + k3
(k1 > 0, k2 < 0, k3 > 0)
(equilibrium)
MD = MS
show that when the money market is in equilibrium
k1Y + k2r = M*S − k3
(3) (a)By solving the simultaneous equations derived in parts (1) and (2), show that
when the commodity and money markets are both in equilibrium
Y5
k2(b 1 d) 1 c(M*S 2 k3)
(1 2 a)k2 1 ck1
(b) Write down the money supply multiplier, ∂Y/∂M*S, and deduce that an increase
in M*S causes an increase in Y.
➜
432 Chapter 5 Partial Differentiation
7. Consider the three-sector model
Y=C+I
C = aYd + b
(0 < a < 1, b > 0)
Yd = Y − T
T = tY + T *
(0 < t < 1, T* > 0)
I = cr + d
(c < 0, d > 0)
(a) Show that
Y5
b 1 d 2 aT* 1 cr
1 2 a(1 2 t)
(b) Find expressions for the multipliers ∂Y/∂c and ∂Y/∂a.
(c) State whether the value of Y increases or decreases as a result of an increase in the
value of c. Give a reason for your answer.
(d) If a = 0.8, b = 100, t = 0.25, T * = 250, c = −60, d = 1700 and r = 8, calculate the
equilibrium level of income, Y, and use the multiplier obtained in part (b) to estimate the change in Y due to a 0.01 increase in the marginal propensity to consume.
8. The demand and total cost functions of a good are P = a − bQ and TC = f + vQ,
respectively, where a, b, f and v are positive constants.
(a 2 v)2
2 f.
4b
−p −p −p
−p
(b) Find expressions for the multipliers,
, ,
and .
−a −b −f
−v
(a) Show that the maximum profit is given by p 5
(c) Use your answers to part (b) to comment on the direction of change in the maximum
profit due to an increase in a, b, f or v.
Section 5.4
Unconstrained optimisation
Objectives
At the end of this section you should be able to:
●
Use the first-order partial derivatives to find the stationary points of a function of
two variables.
●
Use the second-order partial derivatives to classify the stationary points of a function of two
variables.
●
Find the maximum profit of a firm that produces two goods.
●
Find the maximum profit of a firm that sells a single good in different markets with price
discrimination.
As you might expect, methods for finding the maximum and minimum points of a function
of two variables are similar to those used for functions of one variable. However, the nature
of economic functions of several variables forces us to subdivide optimisation problems into
two types, unconstrained and constrained. To understand the distinction, consider the utility
function
3/4
U(x1, x2) = x 1/4
1 x2
The value of U measures the satisfaction gained from buying x1 items of a good G1 and x2
items of a good G2. The natural thing to do here is to try to pick x1 and x2 to make U as large
as possible, thereby maximising utility. However, a moment’s thought should convince you
that, as it stands, this problem does not have a finite solution. The factor x 1/4
1 can be made as
large as we please by taking ever-increasing values of x1 and likewise for the factor x 23/4. In
other words, utility increases without bound as more and more items of goods G1 and G2
are bought. In practice, of course, this does not occur, since there is a limit to the amount of
money that an individual has to spend on these goods. For example, suppose that the cost of
each item of G1 and G2 is $2 and $3, respectively, and that we allocate $100 for the purchase
of these goods. The total cost of buying x1 items of G1 and x2 items of G2 is
2x1 + 3x2
so we require
2x1 + 3x2 = 100
The problem now is to maximise the utility function
3/4
U = x 1/4
1 x2
subject to the budgetary constraint
2x1 + 3x2 = 100
The constraint prevents us from taking ever-increasing values of x1 and x2 and leads to a finite
solution.
434 Chapter 5 Partial Differentiation
We describe how to solve constrained optimisation problems in the following two sections. For the moment, we concentrate on the simple case of optimising functions
z = f (x, y)
without any constraints. This is typified by the problem of profit maximisation, which usually has a finite solution without the need to impose constraints. In a sense the constraints are
built into the profit function, which is defined by
π = TR − TC
because there is a conflict between trying to make total revenue, TR, as large as possible and
trying to make total cost, TC, as small as possible.
Let us begin by recalling how to find and classify stationary points of functions of one variable
y = f (x)
In Section 4.6 we used the following strategy:
Step 1
Solve the equation
f ′(x) = 0
to find the stationary points, x = a.
Step 2
If
●
f ″(a) > 0, then the function has a minimum at x = a
●
f ″(a) < 0, then the function has a maximum at x = a
●
f ″(a) = 0, then the point cannot be classified using the available information.
For functions of two variables
z = f (x, y)
the stationary points are found by solving the simultaneous equations
−z
50
−x
−z
50
−y
that is,
fx(x, y) = 0
fy(x, y) = 0
This is a natural extension of the one-variable case. We first write down expressions for the
first-order partial derivatives and then equate to zero. This represents a system of two equations
for the two unknowns x and y, which we hope can be solved. Stationary points obtained in this
way can be classified into one of three types: minimum point, maximum point and saddle point.
Figure 5.9(a) shows the shape of a surface in the neighbourhood of a minimum. It can be
thought of as the bottom of a bowl-shaped valley. If you stand at the minimum point and walk
in any direction, then you are certain to start moving upwards. Mathematically, we can classify
a stationary point (a, b) as a minimum provided that all three of the following conditions hold:
−2 z
−2z
. 0, 2 . 0,
2
−y
−x
− 2z
−x2
− 2z
− 2z
2
2
−y
−x−y
2
.0
Section 5.4 Unconstrained optimisation 435
(a)
(b)
(c)
Figure 5.9
when x = a and y = b: that is,
fxx(a, b) > 0, fyy(a, b) > 0, fxx(a, b) fyy(a, b) − [ fxy(a, b)]2 > 0
This triple requirement is obviously more complicated than the single condition needed in the
case of a function of one variable. However, once the second-order partial derivatives have
been evaluated at the stationary point, the three conditions are easily checked.
Figure 5.9(b) shows the shape of a surface in the neighbourhood of a maximum. It can be
thought of as the summit of a mountain. If you stand at the maximum point and walk in any
direction, then you are certain to start moving downwards. Mathematically, we can classify a
stationary point (a, b) as a maximum provided that all three of the following conditions hold:
−2z
−2z
0,
,
, 0,
−y2
−x2
− 2z
−x2
− 2z
− 2z
2
−y2
−x−y
2
.0
when x = a and y = b: that is,
fxx(a, b) < 0, fyy(a, b) < 0, fxx(a, b) fyy(a, b) − [ fxy(a, b)]2 > 0
Of course, any particular mountain range may well have lots of valleys and summits.
Likewise, a function of two variables can have more than one minimum or maximum.
Figure 5.9(c) shows the shape of a surface in the neighbourhood of a saddle point. As its
name suggests, it can be thought of as the middle of a horse’s saddle. If you sit at this point
and edge towards the head or tail, then you start moving upwards. On the other hand, if you
edge sideways, then you start moving downwards (and will probably fall off!). Mathematically,
we can classify a stationary point (a, b) as a saddle point provided that the following single
condition holds:
−2z
−x2
− 2z
−2z
2
−y2
−x−y
2
,0
when x = a and y = b: that is,
fxx(a, b) fyy(a, b) − [ fxy(a, b)]2 < 0
To summarise, the method for finding and classifying stationary points of a function f(x, y)
is as follows:
Step 1
Solve the simultaneous equations
fx(x, y) = 0
fy(x, y) = 0
to find the stationary points, (a, b).
436 Chapter 5 Partial Differentiation
Step 2
If
●
fxx > 0, fyy > 0 and fxx fyy − fxy2 > 0 at (a, b), then the function has a minimum at (a, b)
●
fxx < 0, fyy < 0 and fxx fyy − fxy2 > 0 at (a, b), then the function has a maximum at (a, b)
●
fxx fyy − fxy2 < 0 at (a, b), then the function has a saddle point at (a, b).
Before we consider examples from economics, we will illustrate the method by finding and
classifying the stationary points of the function
f (x, y) = x3 − 3x + xy2
In order to use steps 1 and 2 we need to find all first- and second-order partial derivatives of
the function
f (x, y) = x3 − 3x + xy2
These are easily worked out as
fx = 3x2 − 3 + y2
fy = 2xy
fxx = 6x
fxy = 2y
fyy = 2x
Step 1
The stationary points are the solutions of the simultaneous equations
fx(x, y) = 0
fy(x, y) = 0
so we need to solve
3x2 − 3 + y2 = 0
2xy = 0
We have solved simultaneous equations on many occasions throughout this text. So far these
have been linear. This time, however, we need to solve a pair of non-linear equations.
Unfortunately, there is no standard method for solving such systems. We have to rely on our
wits in any particular instance. The trick here is to begin with the second equation
2xy = 0
The only way that the product of three numbers can be equal to zero is when one or more
of the individual numbers forming the product are zero. We know that 2 ≠ 0, so either x = 0
or y = 0. We investigate these two possibilities separately:
●
Case 1: x = 0. Substituting x = 0 into the first equation
3x2 − 3 + y2 = 0
gives
−3 + y2 = 0
that is,
y2 = 3
Section 5.4 Unconstrained optimisation 437
There are therefore two possibilities for y to go with x = 0, namely y 5 2 3 and y 5 3 .
Hence (0, 2 3) and (0, 3) are stationary points.
●
Case 2: y = 0. Substituting y = 0 into the first equation
3x2 − 3 + y2 = 0
gives
3x2 − 3 = 0
that is,
x2 = 1
There are therefore two possibilities for x to go with y = 0, namely x = −1 and x = 1.
Hence (−1, 0) and (1, 0) are stationary points.
These two cases indicate that there are precisely four stationary points, (0, 2 3),
(0, 3), (21, 0) and (1, 0) .
Step 2
To classify these points we need to evaluate the second-order partial derivatives
fxx = 6x, fyy = 2x, fxy = 2y
at each point and check the signs of
fxx, fyy and fxx fyy − f xy2
●
Point (0, 2 3)
fxx = 6(0) = 0, fyy = 2(0) = 0, fxy = 22 3
Hence
fxx fyy 2 f xy2 5 0(0) 2 (22 3)2 5 212 , 0
and so (0, 2 3) is a saddle point.
●
Point (0, 3 )
fxx 5 6(0) 5 0, fyy 5 2(0) 5 0,
fxy 5 2 3
Hence
fxx fyy 2 fxy2 5 0(0) 2 (2 3)2 5 212 , 0
and so (0, 3) is a saddle point.
●
Point (−1, 0)
fxx = 6(−1) = −6, fyy = 2(−1) = −2, fxy = 2(0) = 0
Hence
fxx fyy − f xy2 = (−6)(−2) − 02 = 12 > 0
and so (−1, 0) is not a saddle point. Moreover, since
fxx < 0 and fyy < 0
we deduce that (−1, 0) is a maximum.
438 Chapter 5 Partial Differentiation
●
Point (1, 0)
fxx = 6(1) = 6, fyy = 2(1) = 2, fxy = 2(0) = 0
Hence
fxx fyy − f xy2 = 6(2) − 02 = 12 > 0
and so (1, 0) is not a saddle point. Moreover, since
fxx > 0 and fyy > 0
we deduce that (1, 0) is a minimum.
Practice Problem
1. Find and classify the stationary points of the function
f(x, y) = x2 + 6y − 3y2 + 10
We now consider two examples from economics, both involving the maximisation of
profit. The first considers the case of a firm producing two different goods, whereas the
second involves a single good sold in two different markets.
Example
A firm is a perfectly competitive producer and sells two goods G1 and G2 at $1000 and
$800, respectively. The total cost of producing these goods is given by
TC = 2Q 21 + 2Q1Q2 + Q 22
where Q1 and Q2 denote the output levels of G1 and G2, respectively. Find the maximum
profit and the values of Q1 and Q2 at which this is achieved.
Solution
The fact that the firm is perfectly competitive tells us that the price of each good is fixed
by the market and does not depend on Q1 and Q2. The actual prices are stated in the question as $1000 and $800. If the firm sells Q1 items of G1 priced at $1000, then the revenue is
TR1 = 1000Q1
Similarly, if the firm sells Q2 items of G2 priced at $800, then the revenue is
TR2 = 800Q2
The total revenue from the sale of both goods is then
TR = TR1 + TR2 = 1000Q1 + 800Q2
We are given that the total cost is
TC = 2Q21 + 2Q1Q2 + Q 22
Section 5.4 Unconstrained optimisation 439
so the profit function is
π = TR − TC
= (1000Q1 + 800Q2) − (2Q 21 + 2Q1Q2 + Q 22)
= 1000Q1 + 800Q2 − 2Q 21 − 2Q1Q2 − Q 22
This is a function of the two variables, Q1 and Q2, that we wish to optimise. The first- and
second-order partial derivatives are
−p
5 1000 2 4Q1 2 2Q 2
−Q1
−p
5 800 2 2Q1 2 2Q 2
−Q2
−2p
5 24
−Q12
−2p
5 22
−Q1−Q2
−2p
5 22
−Q22
The two-step strategy then gives the following:
Step 1
At a stationary point
−p
50
−Q1
−p
50
−Q2
so we need to solve the simultaneous equations
1000 − 4Q1 − 2Q2 = 0
800 − 2Q1 − 2Q2 = 0
that is,
4Q1 + 2Q2 = 1000
(1)
2Q1 + 2Q2 = 800
(2)
The variable Q2 can be eliminated by subtracting equation (2) from (1) to get
2Q1 = 200
and so Q1 = 100. Substituting this into either equation (1) or (2) gives Q2 = 300. The profit
function therefore has one stationary point at (100, 300).
➜
440 Chapter 5 Partial Differentiation
Step 2
To show that the point really is a maximum, we need to check that
−2 p
−2 p
−2 p
, 0,
, 0,
2
2
−Q 2
−Q12
−Q 1
−2 p
−2 p
.0
2
2
−Q 2
−Q1−Q2
at this point. In this example the second-order partial derivatives are all constant. We have
−2p
5 24 , 0
−Q 21
✓
−2p
5 22 , 0
−Q 22
✓
−2p
−Q21
−2p
−2p
5 (24)(22) 2 (22)2 5 4 . 0
2
2
−Q2
−Q1−Q2
✓
confirming that the firm’s profit is maximised by producing 100 items of G1 and 300 items
of G2.
The actual value of this profit is obtained by substituting Q1 = 100 and Q2 = 300 into
the expression
π = 1000Q1 + 800Q2 − 2Q 21 − 2Q1Q2 − Q 22
to get
π = 1000(100) + 800(300) − 2(100)2 − 2(100)(300) − (300)2 = $170 000
Practice Problem
2. A firm is a monopolistic producer of two goods G1 and G2. The prices are related to
quantities Q1 and Q2 according to the demand functions
P1 = 50 − Q1
P2 = 95 − 3Q2
If the total cost function is
TC = Q21 + 3Q1Q2 + Q22
show that the firm’s profit function is
π = 50Q1 − 2Q21 − 95Q2 − 4Q22 − 3Q1Q2
Hence find the values of Q1 and Q2 which maximise π and deduce the corresponding
prices.
Section 5.4 Unconstrained optimisation 441
Example
A firm is allowed to charge different prices for its domestic and industrial customers. If
P1 and Q1 denote the price and demand for the domestic market, then the demand
equation is
P1 + Q1 = 500
If P2 and Q2 denote the price and demand for the industrial market, then the demand
equation is
2P2 + 3Q2 = 720
The total cost function is
TC = 50 000 + 20Q
where Q = Q1 + Q2. Determine the firm’s pricing policy that maximises profit with price
discrimination and calculate the value of the maximum profit.
Solution
The topic of price discrimination has already been discussed in Section 4.7. This particular
problem is identical to the worked example solved in that section using ordinary differenti­
ation. You might like to compare the details of the two approaches.
Our current aim is to find an expression for profit in terms of Q1 and Q2 which can then be
optimised using partial differentiation. For the domestic market the demand equation is
P1 + Q1 = 500
which rearranges as
P1 = 500 − Q1
The total revenue function for this market is then
TR1 = P1Q1 = (500 − Q1)Q1 = 500Q1 − Q21
For the industrial market the demand equation is
2P2 + 3Q2 = 720
which rearranges as
P2 = 360 − 3/2Q2
The total revenue function for this market is then
TR2 = P2Q2 = (360 − 3/2 Q2)Q2 = 360Q2 − 3/2 Q22
The total revenue received from sales in both markets is
TR = TR1 + TR2 = 500Q1 − Q21 + 360Q2 − 3/2 Q22
The total cost of producing these goods is given by
TC = 50 000 + 20Q
➜
442 Chapter 5 Partial Differentiation
and, since Q = Q1 + Q2, we can write this as
TC = 50 000 + 20(Q1 + Q2)
= 50 000 + 20Q1 + 20Q2
The firm’s profit function is therefore
π = TR − TC
= (500Q1 − Q21 + 360Q2 − 3/2Q22) − (50 000 + 20Q1 + 20Q2)
480Q1 − Q21 + 340Q2 − 3/2Q22 − 50 000
This is a function of the two variables, Q1 and Q2, that we wish to optimise. The first- and
second-order partial derivatives are
−p
5 480 2 2Q1
−Q1
−p
5 340 2 3Q2
−Q2
−2 p
5 22
−Q 21
−2 p
50
−Q1−Q2
−2 p
5 23
−Q 22
The two-step strategy gives the following:
Step 1
At a stationary point
−p
50
−Q1
−p
50
−Q 2
so we need to solve the simultaneous equations
480 − 2Q1 = 0
340 − 3Q2 = 0
These are easily solved because they are ‘uncoupled’. The first equation immediately gives
Q1 5
480
5 240
2
while the second gives
Q2 5
340
3
Section 5.4 Unconstrained optimisation 443
Step 2
It is easy to check that the conditions for a maximum are satisfied:
−2 p
5 22 , 0
−Q 21
−2 p
5 23 , 0
−Q 22
−2 p
−Q12
−2 p
−2 p
2
2
−Q2
−Q1−Q2
2
5 (22)(23) 2 0 2 5 6 . 0
The question actually asks for the optimum prices rather than the quantities. These are
found by substituting
Q1 5 240 and Q 2 5
340
3
into the corresponding demand equations. For the domestic market
P1 = 500 − Q1 = 500 − 240 = $260
For the industrial market
P2 5 360 2
3
3 340
Q1 5 360 2
5 $190
2
2 3
Finally, we substitute the values of Q1 and Q2 into the profit function
π = 480Q1 − Q21 + 340Q2 − 3/2Q22 − 50 000
to deduce that the maximum profit is $26 866.67.
Practice Problem
3. A firm has the possibility of charging different prices in its domestic and foreign mar-
kets. The corresponding demand equations are given by
Q1 = 300 − P1
Q2 = 400 − 2P2
The total cost function is
TC = 5000 + 100Q
where Q = Q1 + Q2. Determine the prices that the firm should charge to maximise
profit with price discrimination and calculate the value of this profit.
[You have already solved this particular example in Practice Problem 2(a) of Section 4.7.]
444 Chapter 5 Partial Differentiation
Key Terms
Maximum point (of a function of two variables) A point on a surface which has the highest
function value in comparison with other values in its neighbourhood; at such a point the surface
looks like the top of a mountain.
Minimum point (of a function of two variables) A point on a surface which has the lowest function value in comparison with other values in its neighbourhood; at such a point the surface looks
like the bottom of a valley or bowl.
Saddle point A stationary point which is neither a maximum or minimum and at which the surface
looks like the middle of a horse’s saddle.
Exercise 5.4
1. (a) Find the first-order partial derivatives,
−z −z
, , of the function
−x −y
z = 2x2 + y2 − 12x − 8y + 50
and hence find the stationary point.
(b) Find the second-order partial derivatives,
the stationary point is a minimum.
−2z −2z
−2z
, 2 and
, and hence show that
2
−x −y
−x−y
2. Find and classify the stationary points of the following functions:
(a) f (x, y) = x3 + y3 − 3x − 3y (b) f (x, y) = x3 + 3xy2 − 3x2 − 3y2 + 10
3. A firm’s profit function for the production of two goods is given by
π = 24Q1 − Q21 − Q1Q2 − 2Q22 + 33Q2 − 43
Find the output levels needed to maximise profit. Use second-order derivatives to confirm that the stationary point is a maximum.
4. A perfectly competitive producer sells two goods, G1 and G2, at $70 and $50, respec-
tively. The total cost of producing these goods is given by
TC = Q21 + Q1Q2 + Q22
where Q1 and Q2 denote the output levels of G1 and G2. Find the maximum profit and
the values of Q1 and Q2 at which this is achieved.
5. An individual’s utility function is given by
U = 260x1 + 310x2 + 5x1x2 − 10x21 − x22
where x1 is the amount of leisure measured in hours per week and x2 is earned income
measured in dollars per week. Find the values of x1 and x2 which maximise U. What is
the corresponding hourly rate of pay?
6. A monopolist produces the same product at two factories. The cost functions for each
factory are as follows:
TC1 = 8Q1
and TC2 = Q22
The demand function for the good is
P = 100 − 2Q
where Q = Q1 + Q2. Find the values of Q1 and Q2 which maximise profit.
Section 5.4 Unconstrained optimisation 445
7. A monopolist sells its product in two isolated markets with demand functions
P1 = 32 − Q1
and P2 = 40 − 2Q2
The total cost function is TC = 4(Q1 + Q2).
(a) Show that the profit function is given by
π = 28Q1 + 36Q2 − Q21 − 2Q22
(b) Find the values of Q1 and Q2 which maximise profit and calculate the value of the
maximum profit. Verify that the second-order conditions for a maximum are satisfied.
8. (a) If the monopolist in Question 7 is no longer allowed to discriminate between the two
markets and must charge the same price, P, show that the total demand, Q = Q1 + Q2,
is given by
Q 5 52 2
3
P
2
1
3
and deduce that the profit function is p 5 (92Q 2 2Q2).
(b) Find the maximum profit in part (a) and hence work out how much profit is lost by
the firm when it can no longer discriminate between the two markets.
Exercise 5.4*
1. Find and classify the stationary points of the each of the following functions:
(a) f (x, y) = x3 + x2 − xy + y2 + 10
(b) f (x, y) = (2xy + y2)ex
(c) f(x, y) = x2 − y2 − 4xy − y3
2. A firm’s production function is given by
Q = 2L1/2 + 3K1/2
where Q, L and K denote the number of units of output, labour and capital. Labour
costs are $2 per unit, capital costs are $1 per unit and output sells at $8 per unit. Show
that the profit function is
π = 16L1/2 + 24K1/2 − 2L − K
and hence find the maximum profit and the values of L and K at which it is achieved.
3. A firm sells a good to both UK and EU customers. The demand function is the same
for both markets and is given by
20Pi + Qi = 5000
where the subscript, i, takes the values 1 and 2 corresponding to the UK and EU, respectively.
Although the variable and fixed costs are the same for each market, the EU now
charges a fixed tariff of $50 per unit, so the joint total cost function is
TC = 40Q1 + 90Q2 + 2000
Find the maximum total profit.
➜
446 Chapter 5 Partial Differentiation
4. The demand functions for a firm’s domestic and foreign markets are
P1 = 50 − 5Q1
P2 = 30 − 4Q2
and the total cost function is
TC = 10 + 10Q
where Q = Q1 + Q2. Determine the prices needed to maximise profit with price discrimination and calculate the value of the maximum profit.
[You have already solved this particular example in Question 3(a) in Exercise 4.7*.]
5. A firm is able to sell its product in two different markets. The corresponding demand
functions are
P1 + 2Q1 = 100
2P2 + Q2 = 2a
and the total cost function is
TC = 500 + 10Q
where Q = Q1 + Q2 and a is a positive constant.
Determine, in terms of a, the prices needed to maximise profit
(a) with price discrimination;
(b) without price discrimination.
Show that the profit with price discrimination is always greater than the profit without
discrimination, irrespective of the value of a.
6. (a) A firm’s production function is given by Q = f (K, L). The fixed prices of each unit
of output, capital and labour are p, r and w, respectively. Explain briefly why the
profit function is given by
π(K, L) = pf (K, L) − rK − wL
Show that at a stationary point
−f
r
5
−K
p
and
w
−f
5
−L
p
Give an economic interpretation of these results.
(b) Use the result of part (a), or otherwise, to find the values of K and L which maximise
profit when the production function is Q = K1/2L1/3, p = 24, r = 1.5 and w = 8. Verify
that the second-order conditions for a maximum are satisfied in this case.
7. A firm manufactures two goods labelled 1 and 2. It sells Qi items of good i for a fixed
price per unit of pi. The total cost of producing good i is ciQ 2i .
Explain briefly why the profit function is given by
π(Q1, Q2) = p1Q1 + p2Q2 − c1Q21 − c2Q22
Find the values of Q1 and Q2 which maximise π and verify that the second-order conditions for a maximum are satisfied. Find an expression for the maximum profit.
8. The unit prices of two goods A and B are p and q. The total cost of producing x items
of type A and y items of type B is
TC = 4x2 + 2y2 + 2xy + 100
Find the values of x and y that maximise profit.
Section 5.5
Constrained optimisation
Objectives
At the end of this section you should be able to:
●
Give a graphical interpretation of constrained optimisation.
●
Show that when a firm maximises output subject to a cost constraint, the ratio of marginal
product to price is the same for all inputs.
●
Show that when a consumer maximises utility subject to a budgetary constraint, the ratio of
marginal utility to price is the same for all goods.
●
Use the method of substitution to solve constrained optimisation problems in economics.
Advice
In this section we begin by proving some theoretical results before describing the method of sub­
stitution. You might prefer to skip the theory at a first reading and begin with the examples.
In Section 5.4 we described how to find the optimum (that is, maximum or minimum) of
a function of two variables
z = f(x, y)
where the variables x and y are free to take any values. As we pointed out at the beginning
of that section, this assumption is unrealistic in many economic situations. An individual
wishing to maximise utility is subject to an income constraint, and a firm wishing to maxi­
mise output is subject to a cost constraint.
In general, we want to optimise a function,
z = f(x, y)
called the objective function subject to a constraint
w(x, y) = M
Here w, the Greek letter phi, is a known function of two variables and M is a known constant.
The problem is to pick the pair of numbers (x, y) which maximises or minimises f(x, y) as
before. This time, however, we limit the choice of pairs to those which satisfy
w(x, y) = M
A graphical interpretation should make this clear. To be specific, let us suppose that a firm
wants to maximise output and that the production function is of the form
Q = f(K, L)
448 Chapter 5 Partial Differentiation
Let the costs of each unit of capital and labour be PK and PL, respectively. The cost to the firm
of using as input K units of capital and L units of labour is
PKK + PLL
so if the firm has a fixed amount, M, to spend on these inputs, then
PKK + PLL = M
The problem is one of trying to maximise the objective function
Q = f(K, L)
subject to the cost constraint
PKK + PLL = M
Sketched in Figure 5.10 is a typical isoquant map. As usual, points on any one isoquant yield
the same level of output, and as output rises, the isoquants themselves move further away
from the origin. Also sketched in Figure 5.10 is the cost constraint. This is called an isocost
curve because it gives all combinations of K and L which can be bought for a fixed cost, M.
The fact that
PKK + PLL = M
is represented by a straight line should come as no surprise to you by now. We can even
rewrite it in the more familiar ‘y = ax + b’ form and so identify its slope and intercept. In
Figure 5.10, L is plotted on the horizontal axis and K is plotted on the vertical axis, so we
need to rearrange
PKK + PLL = M
to express K in terms of L. Subtracting PLL from both sides and dividing through by PK gives
K5 2
PL
M
L1
PK
PK
The isocost curve is therefore a straight line with slope –PL/PK and intercept M/PK. Graphically,
our constrained problem is to choose that point on the isocost line which maximises output.
This is given by the point labelled A in Figure 5.10. Point A certainly lies on the isocost line
K
B
A
C
L
Figure 5.10
Section 5.5 Constrained optimisation 449
and it maximises output because it also lies on the highest isoquant. Other points, such as B
and C, also satisfy the constraint, but they lie on lower isoquants and so yield smaller levels
of output than A. Point A is characterised by the fact that the isocost line is tangential to an
isoquant. In other words, the slope of the isocost line is the same as that of the isoquant at A.
Now we have already shown that the isocost line has slope –PL/PK. In Section 5.2 we
defined the marginal rate of technical substitution, MRTS, to be minus the slope of an
isoquant, so at point A we must have
PL
5 MRTS
PK
We also showed that
MRTS 5
MPL
MPK
so
PL
MPL
5
PK MPK
the ratio of the input prices is equal to the ratio of their marginal products
This relationship can be rearranged as
MPL MPK
5
PL
PK
so when output is maximised subject to a cost constraint
the ratio of marginal product to price is the same for all inputs
The marginal product determines the change in output due to a 1-unit increase in input. This
optimisation condition therefore states that the last dollar spent on labour yields the same addition to output as the last dollar spent on capital.
The above discussion has concentrated on production functions. An analogous situation
arises when we maximise utility functions
U = U(x1, x2)
where x1, x2 denote the number of items of goods G1, G2 that an individual buys. If the prices
of these goods are denoted by P1 and P2 and the individual has a fixed budget, M, to spend
on these goods then, the corresponding constraint is
P1x1 + P2x2 = M
This budgetary constraint plays the role of the cost constraint, and indifference curves are
analogous to isoquants. Consequently, we analyse the problem by superimposing the budget
line on an indifference map. The corresponding diagram is virtually indistinguishable from that
of Figure 5.10. The only change is that the axes would be labelled x1 and x2 rather than L and
K. Once again, the maximum point of the constrained problem occurs at the point of tangency,
so that at this point the slope of the budget line is that of an indifference curve. Hence
P1
5 MRCS
P2
450 Chapter 5 Partial Differentiation
In Section 5.2 we derived the result
MRCS 5
−U/−x1
−U/−x2
Writing the partial derivatives ∂U/∂xi more concisely as Ui we can deduce that
P1
U
5 1
P2
U2
that is,
the ratio of the prices of the goods is equal to the ratio of their marginal utilities
Again, this relationship can be rearranged into the more familiar form
U1
U
5 2
P1
P2
so when utility is maximised subject to a budgetary constraint,
the ratio of marginal utility to price is the same for all goods consumed
If individuals allocate their budgets between goods in this way, then utility is maximised
when the last dollar spent on each good yields the same addition to total utility. Under these
circumstances, the consumer has achieved maximum satisfaction within the constraint of a
fixed budget, so there is no tendency to reallocate income between these goods. Obviously,
the consumer’s equilibrium will be affected if there is a change in external conditions such
as income or the price of any good. For example, suppose that P1 suddenly increases, while
P2 and M remain fixed. If this happens, then the equation
U1
U
5 2
P1
P2
turns into an inequality
U1
U
, 2
P1
P2
so equilibrium no longer holds. Given that P1 has increased, consumers find that the last
dollar spent no longer buys as many items of G1, so utility can be increased by purchasing
more of G2 and less of G1. By the law of diminishing marginal utility, the effect is to increase
U1 and to decrease U2. The process of reallocation continues until the ratio of marginal util­
ities to prices is again equal and equilibrium is again established.
The graphical approach provides a useful interpretation of constrained optimisation.
It has also enabled us to justify some familiar results in microeconomics. However, it does
not give us a practical way of actually solving such problems. It is very difficult to produce
an accurate isoquant or indifference map from any given production or utility function. We
now describe an alternative approach, known as the method of substitution.
To illustrate the method we will show how to find the minimum value of the objective
function
z = −2x2 + y2
subject to the constraint y = 2x – 1.
Section 5.5 Constrained optimisation 451
The obvious thing to do is to substitute the expression for y given by the constraint directly
into the function that we are trying to optimise to get
z = −2x2 + (2x − 1)2
= −2x2 + 4x2 − 4x + 1
= 2x2 − 4x + 1
Note the wonderful effect that this has on z. Instead of z being a function of two variables, x
and y, it is now just a function of the one variable, x. Consequently, the minimum value of z
can be found using the theory of stationary points discussed in Chapter 4.
At a stationary point
dz
50
dx
that is,
4x − 4 = 0
which has solution x = 1. Differentiating a second time, we see that
d 2z
54.0
dx
confirming that the stationary point is a minimum. The value of z can be found by substituting
x = 1 into
z = 2x 2 − 4x + 1
to get
z = 2(1)2 − 4(1) + 1 = −1
It is also possible to find the value of y at the minimum. To do this we substitute x = 1 into
the constraint
y = 2x − 1
to get
y = 2(1) − 1 = 1
The constrained function therefore has a minimum value of –1 at the point (1, 1).
The method of substitution for optimising
z = f(x, y)
subject to
w(x, y) = M
may be summarised as follows:
Step 1
Use the constraint
w(x, y) = M
to express y in terms of x.
452 Chapter 5 Partial Differentiation
Step 2
Substitute this expression for y into the objective function
z = f(x, y)
to write z as a function of x only.
Step 3
Use the theory of stationary points of functions of one variable to optimise z.
Practice Problem
1. Find the maximum value of the objective function
z = 2x2 − 3xy + 2y + 10
subject to the constraint y = x.
Advice
The most difficult part of the three-step strategy is step 1, where we rearrange the given constraint
to write y in terms of x. In Practice Problem 1 this step was exceptionally easy because the constraint
was linear. The constraint was even presented in the appropriate form to begin with, so no extra
work was required. In general, if the constraint is non-linear, it may be difficult or impossible to perform the initial rearrangement. If this happens, then you could try working the other way round and
expressing x in terms of y, although there is no guarantee that this will be possible, either. However,
when step 1 can be tackled successfully, the method does provide a really quick way of solving constrained optimisation problems.
To illustrate this we now use the method of substitution to solve two economic problems
that both involve production functions. In the first example, output is maximised subject to
cost constraint, and in the second example, cost is minimised subject to an output constraint.
Example
A firm’s unit capital and labour costs are $1 and $2, respectively. If the production function
is given by
Q = 4LK + L2
find the maximum output and the levels of K and L at which it is achieved when the total
input costs are fixed at $105. Verify that the ratio of marginal product to price is the same
for both inputs at the optimum.
Solution
We are told that 1 unit of capital costs $1 and that 1 unit of labour costs $2. If the firm uses
K units of capital and L units of labour, then the total input costs are
K + 2L
Section 5.5 Constrained optimisation 453
This is fixed at $105, so
K + 2L = 105
The mathematical problem is to maximise the objective function
Q = 4LK + L2
subject to the constraint
K + 2L = 105
The three-step strategy is as follows:
Step 1
Rearranging the constraint to express K in terms of L gives
K = 105 − 2L
Step 2
Substituting this into the objective function
Q = 4LK + L2
gives
Q = 4L(105 − 2L) + L2 = 420L − 7L2
and so output is now a function of the one variable, L.
Step 3
At a stationary point
dQ
50
dL
that is,
420 − 14L = 0
which has solution L = 30. Differentiating a second time gives
d 2Q
5 214 , 0
dL2
confirming that the stationary point is a maximum.
The maximum output is found by substituting L = 30 into the objective function
Q = 420L − 7L2
to get
Q = 420(30) − 7(30)2 = 6300
The corresponding level of capital is found by substituting L = 30 into the constraint
K = 105 − 2L
to get
K = 105 − 2(30) = 45
The firm should therefore use 30 units of labour and 45 units of capital to produce a
maximum output of 6300.
➜
454 Chapter 5 Partial Differentiation
Finally, we are asked to check that the ratio of marginal product to price is the same for
both inputs. From the formula
Q = 4LK + L2
we see that the marginal products are given by
MPL 5
−Q
5 4K 1 2L
−L
and MPK 5
−Q
5 4L
−K
so at the optimum
MPL = 4(45) + 2(30) = 240
and
MPK = 4(30) = 120
The ratios of marginal products to prices are then
240
MPL
5 120
5
PL
2
and
120
MPK
5 120
5
PK
1
which are seen to be the same.
Practice Problem
2. An individual’s utility function is given by
U = x1x2
where x1 and x2 denote the number of items of two goods, G1 and G2. The prices of
the goods are $2 and $10, respectively. Assuming that the individual has $400 available
to spend on these goods, find the utility-maximising values of x1 and x2. Verify that the
ratio of marginal utility to price is the same for both goods at the optimum.
Example
A firm’s production function is given by
Q = 2K1/2L1/2
Unit capital and labour costs are $4 and $3, respectively. Find the values of K and L which
minimise total input costs if the firm is contracted to provide 160 units of output.
Section 5.5 Constrained optimisation 455
Solution
Given that capital and labour costs are $4 and $3 per unit, the total cost of using K units
of capital and L units of labour is
TC = 4K + 3L
The firm’s production quota is 160, so
2K1/2L1/2 = 160
The mathematical problem is to minimise the objective function
TC = 4K + 3L
subject to the constraint
2K1/2L1/2 = 160
Step 1
Rearranging the constraint to express L in terms of K gives
L 1/2 5
L5
80
K 1/2
(divide both sides by 2K1/2)
6400
K
(square both sides)
Step 2
Substituting this into the objective function
TC = 4K + 3L
gives
TC 5 4K 1
19 200
K
and so total cost is now a function of the one variable, K.
Step 3
At a stationary point
d(TC)
50
dK
that is,
42
19 200
50
K2
This can be written as
45
19 200
K2
➜
456 Chapter 5 Partial Differentiation
so that
K2 5
19 200
5 4800
4
Hence
K 5 4800 5 69.28
Differentiating a second time gives
d 2(TC) 38 400
. 0 because K . 0
5
dK 2
K3
confirming that the stationary point is a minimum.
Finally, the value of L can be found by substituting K = 69.28 into the constraint
L5
6400
K
to get
L5
6400
5 92.38
69.28
We are not asked for the minimum cost, although this could easily be found by substituting the values of K and L into the objective function.
Practice Problem
3. A firm’s total cost function is given by
TC = 3x 21 + 2x1x2 + 7x 22
where x1 and x2 denote the number of items of goods G1 and G2, respectively, that are
produced. Find the values of x1 and x2 which minimise costs if the firm is committed to
providing 40 goods of either type in total.
Key Terms
Isocost curve
A line showing all combinations of two factors which can be bought for a fixed cost.
Method of substitution The method of solving constrained optimisation problems whereby the
constraint is used to eliminate one of the variables in the objective function.
Objective function A function that one seeks to optimise (usually) subject to constraints.
Section 5.5 Constrained optimisation 457
Exercise 5.5
1. (a) Make y the subject of the formula 9x + 3y = 2.
(b) The function,
z = 3xy
is subject to the constraint
9x + 3y = 2
Use your answer to part (a) to show that
z = 2x − 9x2
Hence find the maximum value of z and the corresponding values of x and y.
2. Find the maximum value of
z = 6x − 3x2 + 2y
subject to the constraint
y − x2 = 2
3. Find the maximum value of
z = 80x − 0.1x2 + 100y − 0.2y2
subject to the constraint
x + y = 500
4. A firm’s production function is given by
Q = 50KL
Unit capital and labour costs are $2 and $3, respectively. Find the values of K and L
which minimise total input costs if the production quota is 1200.
5. The total cost of producing x items of product A and y items of product B is
TC = 22x2 + 8y2 − 5xy
If the firm is committed to producing 20 items in total, write down the constraint
connecting x and y. Hence find the number of each type that should be produced to
minimise costs.
6. Find the maximum value of the utility function, U = x1x2, subject to the budgetary
constraint, x1 + 4x2 = 360.
7. A firm produces two goods A and B. The weekly cost of producing x items of A and y
items of B is
TC = 0.2x2 + 0.05y2 + 0.1xy + 2x + 5y + 1000
(a) State the minimum value of TC in the case when there are no constraints.
(b) Find the minimum value of TC when the firm is committed to producing 500 goods
of either type in total.
➜
458 Chapter 5 Partial Differentiation
8. A firm’s production function is given by Q = AKL where A is a positive constant.
Unit costs of capital and labour are $2 and $1, respectively. The firm spends a total of
$1000 on these inputs.
(a) Write down the cost constraint.
(b) Write down expressions for the marginal products of capital and labour.
(c) Use the fact that at the maximum output, the ratio of the input prices is equal to
the ratio of their marginal products to show that L = 2K.
(d) Use your answers to parts (a) and (c) to find the values of K and L which maximise
output.
Exercise 5.5*
1. (a) Find the minimum value of the objective function
z = 9x2 + 2y2 − 3xy
subject to the constraint
x + y = 40.
(b) Find the maximum value of the objective function
−16x2 − 2y2 + 4x + 9y + 2xy
subject to the constraint
y = 4x
2. A firm’s production function is given by
Q = 10K1/2L1/4
Unit capital and labour costs are $4 and $5, respectively, and the firm spends a total of
$60 on these inputs. Find the values of K and L which maximise output.
3. A firm’s production function is given by
Q = 2L1/2 + 3K1/2
where Q, L and K denote the number of units of output, labour and capital, respectively.
Labour costs are $2 per unit, capital costs are $1 per unit and output sells at $8 per unit.
If the firm is prepared to spend $99 on input costs, find the maximum profit and the
values of K and L at which it is achieved.
[You might like to compare your answer with the corresponding unconstrained problem that you solved in Question 2 of Exercise 5.4*.]
4. A consumer’s utility function is
U = ln x1 + 2 ln x2
Find the values of x1 and x2 which maximise U subject to the budgetary constraint
2x1 + 3x2 = 18
Section 5.5 Constrained optimisation 459
5. An individual’s utility function is given by
U 5 x1 x2
where x1 and x2 denote the monthly consumption of two goods. Unit prices of these
goods are $2 and $4, respectively, and the total monthly expenditure on these goods is
$300.
(a) Write down the budgetary constraint.
(b) Show that if the budgetary constraint is satisfied, the maximum value of U is 500,
and write down the corresponding values of x1 and x2. Verify that the stationary
point is a maximum.
(c) Draw the three indifference curves, x1 x2 5 400, x1 x2 5 500 and x1 x2 5 600, on
the same diagram. Hence show that the maximum point of the constrained problem
occurs at a point where the budgetary constraint is a tangent to an indifference
curve.
6. A firm wishes to maximise output, Q = K2L, subject to a budgetary constraint
3K + 2L = 900
(a) Find the values of K and L which maximise output.
(b) Draw a graph to illustrate the fact that at this maximum output the budget line is a
tangent to a particular isoquant.
7. A utility function, U(x1, x2) 5 x1 1 x2, is subject to a budgetary constraint, x2 = b − ax1
where a and b are positive constants.
(a) Show that the maximum value of U is U* 5
(b) By considering the signs of
4ab 1 1
.
4a
−U*
−U*
and
, state what happens to the value of the
−a
−b
optimal utility due to changes in a or b.
Section 5.6
Lagrange multipliers
Objectives
At the end of this section you should be able to:
●
Use the method of Lagrange multipliers to solve constrained optimisation problems.
●
Give an economic interpretation of Lagrange multipliers.
●
Use Lagrange multipliers to maximise a Cobb–Douglas production function subject to a cost
constraint.
●
Use Lagrange multipliers to show that when a firm maximises output subject to a cost
constraint, the ratio of marginal product to price is the same for all inputs.
We now describe the method of Lagrange multipliers for solving constrained optimisation
problems. This is the preferred method, since it handles non-linear constraints and problems
involving more than two variables with ease. It also provides some additional information
that is useful when solving economic problems.
To optimise an objective function
f(x, y)
subject to a constraint
w(x, y) = M
we work as follows.
Step 1
Define a new function
g(x, y, l) = f(x, y) + l[M − w(x, y)]
Step 2
Solve the simultaneous equations
−g
50
−x
−g
50
−y
−g
50
−l
for the three unknowns, x, y and l.
The basic steps of the method are straightforward. In step 1 we combine the objective
function and constraint into a single function. To do this we first consider
M − w(x, y)
Section 5.6 Lagrange multipliers 461
and multiply by the scalar (i.e. number) l (the Greek letter ‘lambda’). This scalar is called the
Lagrange multiplier. Finally, we add on the objective function to produce the new function
g(x, y, l) = f(x, y) + l[M − w(x, y)]
This is called the Lagrangian function. The right-hand side involves the three letters x, y and
l, so g is a function of three variables.
In step 2 we work out the three first-order partial derivatives
−g −g −g
,
,
−x −y −l
and equate these to zero to produce a system of three simultaneous equations for the three
unknowns x, y and l. The point (x, y) is then the optimal solution of the constrained problem. The number l can also be given a meaning, and we consider this later.
To illustrate the new method consider trying to optimise the value of
x2 − 3xy + 12x
subject to the constraint
2x + 3y = 6
Step 1
In this case
f(x, y) = x2 − 3xy + 12x
w(x, y) = 2x + 3y
M=6
so the Lagrangian function is given by
g(x, y, l) = x2 − 3xy + 12x + l(6 − 2x − 3y)
Step 2
Working out the three partial derivatives of g gives
−g
5 2x 2 3y 1 12 2 2l
−x
−g
5 23x 2 3l
−y
−g
5 6 2 2x 2 3y
−l
so we need to solve the simultaneous equations
2x − 3y + 12 − 2l = 0
−3x − 3l = 0
6 − 2x − 3y = 0
that is,
2x − 3y − 2l = −12
(1)
−3x − 3l = 0
(2)
2x + 3y = 6
(3)
462 Chapter 5 Partial Differentiation
We can eliminate x from equation (2) by multiplying equation (1) by 3, multiplying equation (2)
by 2 and adding. Similarly, x can be eliminated from equation (3) by subtracting equation (3)
from (1). These operations give
−9y − 12l = −36
(4)
−6y − 2l = −18
(5)
The variable y can be eliminated by multiplying equation (4) by 6 and equation (5) by 9, and
subtracting to get
−54l = −54
(6)
so l = 1. Substituting this into equations (5) and (2) gives y = 8/3 and x = –1, respectively.
The optimal solution is therefore (–1, 8/3), and the corresponding value of the objective
function
x2 − 3xy + 12x
is
(−1)2 − 3(–1)(8/3) + 12(−1) = −3
Practice Problem
1. Use Lagrange multipliers to optimise
2x2 − xy
subject to
x + y = 12
Looking back at your own solution to Practice Problem 1, notice that the third equation
in step 2 is just a restatement of the original constraint. It is easy to see that this is always the
case because if
g(x, y, l) = f(x, y) + l[M − w(x, y)]
then
−g
5 M 2 w(x, y)
−l
The equation
−g
50
−l
then implies the constraint
w(x, y) = M
Section 5.6 Lagrange multipliers 463
Example
A monopolistic producer of two goods, G1 and G2, has a joint total cost function
TC = 10Q1 + Q1Q2 + 10Q2
where Q1 and Q2 denote the quantities of G1 and G2 respectively. If P1 and P2 denote the
corresponding prices then the demand equations are
P1 = 50 − Q1 + Q2
P2 = 30 + 2Q1 − Q2
Find the maximum profit if the firm is contracted to produce a total of 15 goods of either
type. Estimate the new optimal profit if the production quota rises by 1 unit.
Solution
The first thing that we need to do is to write down expressions for the objective function
and constraint. The objective function is profit and is given by
π = TR − TC
The total cost function is given to be
TC = 10Q1 + Q1Q2 + 10Q2
However, we need to use the demand equations to obtain an expression for TR. Total
revenue from the sale of G1 is
TR1 = P1Q1 = (50 − Q1 + Q2)Q1 = 50Q1 − Q 21 + Q2Q1
and total revenue from the sale of G2 is
TR2 = P2Q2 = (30 + 2Q1 − Q2)Q2 = 30Q2 + 2Q1Q2 − Q 22
so
TR = TR1 + TR2
= 50Q1 − Q 21 + Q2Q1 + 30Q2 + 2Q1Q2 − Q 22
= 50Q1 − Q 21 + 3Q1Q2 + 30Q2 − Q 22
Hence
π = TR − TC
= (50Q1 − Q 21 + 3Q1Q2 + 30Q2 − Q 22 ) − (10Q1 + Q1Q2 + 10Q2)
= 40Q1 − Q 21 + 2Q1Q2 + 20Q2 − Q 22
The constraint is more easily determined. We are told that the firm produces 15 goods in
total, so
Q1 + Q2 = 15
The mathematical problem is to maximise the objective function
π = 40Q1 − Q 21 + 2Q1Q2 + 20Q2 − Q 22
subject to the constraint
Q1 + Q2 = 15
➜
464 Chapter 5 Partial Differentiation
Step 1
The Lagrangian function is
g(Q1, Q2, l) = 40Q1 − Q 21 + 2Q1Q2 + 20Q2 − Q 22 + l(15 − Q1 − Q2 )
Step 2
The simultaneous equations
−g
−g
−g
50
5 0,
5 0,
−Q1
−Q2
−l
are
40 − 2Q1 + 2Q2 − l = 0
2Q1 + 20 − 2Q2 − l = 0
15 − Q1 − Q2 = 0
that is,
−2Q1 + 2Q2 − l = −40
(1)
2Q1 − 2Q2 − l = −20
(2)
Q1 + Q2 = 15
(3)
The obvious way of solving this system is to add equations (1) and (2) to get
−2l = −60
so l = 30. Putting this into equation (1) gives
−2Q1 + 2Q2 = −10
(4)
Equations (3) and (4) constitute a system of two equations for the two unknowns Q1 and
Q2. We can eliminate Q1 by multiplying equation (3) by 2 and adding equation (4) to get
4Q2 = 20
so Q2 = 5. Substituting this into equation (3) gives
Q1 = 15 − 5 = 10
The maximum profit is found by substituting Q1 = 10 and Q2 = 5 into the formula for π to
get
π = 40(10) − (10)2 + 2(10)(5) + 20(5) − 52 = 475
The final part of this example wants us to find the new optimal profit when the production quota rises by 1 unit. One way of doing this is just to repeat the calculations replacing
the previous quota of 15 by 16, although this is extremely tedious and not strictly necessary. There is a convenient shortcut based on the value of the Lagrange multiplier l. To
understand this, let us replace the production quota, 15, by the variable M, so that the
Lagrangian function is
g(Q1, Q2, l, M) = 40Q1 − Q 21 + 2Q1Q2 + 20Q2 − Q 22 + l(M − Q1 − Q2)
Section 5.6 Lagrange multipliers 465
The expression on the right-hand side involves Q1, Q2, l and M, so g is now a function of
four variables. If we partially differentiate with respect to M, then
−g
5l
−M
We see that l is a multiplier not only in the mathematical sense but also in the economic
sense. It represents the (approximate) change in g due to a 1-unit increase in M. Moreover,
if the constraint is satisfied, then
Q1 + Q2 = M
and the expression for g reduces to
40Q1 − Q 21 + 2Q1Q2 + 20Q2 − Q 22
which is equal to the profit. The value of the Lagrange multiplier represents the change
in optimal profit brought about by a 1-unit increase in the production quota. In this case,
l = 30, so the profit rises by 30 to become 505.
The interpretation placed on the value of l in this example applies quite generally. Given
an objective function
f(x, y)
and constraint
w(x, y) = M
the value of l gives the approximate change in the optimal value of f due to a 1-unit increase
in M.
Practice Problem
2. A consumer’s utility function is given by
U(x1, x2) = 2x1x2 + 3x1
where x1 and x2 denote the number of items of two goods G1 and G2 that are bought.
Each item costs $1 for G1 and $2 for G2. Use Lagrange multipliers to find the max­
imum value of U if the consumer’s income is $83. Estimate the new optimal utility if
the consumer’s income rises by $1.
The Lagrange multiplier method can be used to solve the problem of maximising the general Cobb–Douglas production function
Q = AK aLb
(A, a and b are positive constants)
when it is subject to a cost constraint
PKK + PLL = M
This problem appears very hard at first sight because it does not involve specific numbers.
However, it is easy to handle such generalised problems provided that we do not panic.
466 Chapter 5 Partial Differentiation
Step 1
The Lagrangian function is
g(K, L, l) = AKaLb + l(M − PKK − PLL)
Step 2
The simultaneous equations
−g
−g
−g
5 0,
5 0,
50
−K
−L
−l
are
AaKa−1Lb − lPK = 0
(1)
AbKaLb−1 − lPL = 0
(2)
M − PKK − PLL = 0
(3)
These equations look rather forbidding. Before we begin to solve them, it pays to simplify
equations (1) and (2) slightly by introducing Q = AKaLb. Notice that
AaKa21L b 5
aQ
a(AKaLb)
5
K
K
AbK aLb21 5
b(AK aLb)
bQ
5
L
L
so equations (1), (2) and (3) can be written
aQ
2 lPK 5 0
K
(4)
bQ
2 lPL 5 0
L
(5)
PKK + PLL = M
(6)
Equations (4) and (5) can be rearranged to give
l5
aQ
PK K
and l 5
bQ
PL L
so that
aQ
bQ
5
PK K
PL L
and hence
PK K
PL
5 L
a
b
(divide both sides by Q and turn both sides upside down)
that is,
PK K 5
a
PL
b L
(multiply through by a)
(7)
Section 5.6 Lagrange multipliers 467
Substituting this into equation (6) gives
a
P L 1 PL L 5 M
b L
aL 1 bL 5
bM
PL
(multiply through by b/PL)
(a 1 b)L 5
bM
PL
(factorise)
L5
bM
(a 1 b)PL
(divide through by a 1 b)
Finally, we can put this into equation (7) to get
PK K 5
aM
a1b
so
K5
aM
(a 1 b)PK
The values of K and L which optimise Q are therefore
aM
(a 1 b)PK
and
bM
(a 1 b)PL
Practice Problem
3. Use Lagrange multipliers to find expressions for x1 and x2 which maximise the utility
function
U = x 11/2 + x 21/2
subject to the general budgetary constraint
P1x1 + P2x2 = M
The previous work illustrates the power of mathematics when solving economics problems. The main advantage of using algebra and calculus rather than just graphs and tables of
numbers is their generality. In future, if we need to maximise any particular Cobb–Douglas
production function subject to any particular cost constraint, all we have to do is to quote
the previous result. By substituting specific values of M, a, b, PK and PL into the general formulas for K and L, we can write down the solution in a matter of seconds. In fact, we can use
mathematics to generalise still further. Rather than work with production functions of a
prescribed form such as
Q = AKaLb
we can obtain results pertaining to any production function
Q = f(K, L)
For instance, we can use Lagrange multipliers to justify a result that we derived graphically in
Section 5.5. At the beginning of that section we showed that when output is maximised, subject
468 Chapter 5 Partial Differentiation
to a cost constraint, the ratio of marginal product to price is the same for all inputs. To obtain
this result using Lagrange multipliers we simply write down the Lagrangian function
g(K, L, l) = f(K, L) + l(M − PKK − PLL)
which corresponds to a production function
f(K, L)
and cost constraint
PKK + PLL = M
The simultaneous equations
−g
−g
−g
5 0,
5 0,
50
−K
−L
−l
are
MPK − lPK = 0
(1)
MPL − lPL = 0
(2)
M − PKK − PLL = 0
(3)
because
−f
5 MPK
−K
and
−f
5 MPL
−K
Equations (1) and (2) can be rearranged to give
l5
MPK
PK
and l 5
MPL
PL
so
MPK
MPL
5
PK
PL
as required.
Key Terms
Lagrange multiplier The number l which is used in the Lagrangian function. In economics this
gives the approximate change in the optimal value of the objective function when the value of the
constraint is increased by 1 unit.
Lagrangian The function f(x, y) + l[M − w(x, y)], where f(x, y) is the objective function and
w(x, y) = M is the constraint. The stationary point of this function is the solution of the asso­
ciated constrained optimisation problem.
Section 5.6 Lagrange multipliers 469
Advice
There now follow some additional problems for you to try. If you feel that you need even more practice, then you are advised to rework the questions in Section 5.5 using Lagrange multipliers.
Exercise 5.6
1. Use Lagrange multipliers to find the maximum value of
z = x + 2xy
subject to the constraint
x + 2y = 5
2. (a) Use Lagrange multipliers to find the maximum value of
z = 4xy
subject to the constraint
x + 2y = 40
State the associated values of x, y and l.
(b) Repeat part (a) when the constraint is changed to
x + 2y = 41
(c) Verify that the value of the Lagrange multiplier in part (a) is approximately the
same as the change in the optimal value of z when the right-hand side of the constraint is increased by 1 unit.
3. A firm’s production function is given by
Q = KL
Unit capital and labour costs are $2 and $1, respectively. Find the maximum level of
output if the total cost of capital and labour is $6.
4. A firm’s production function is given by Q = 80KL. Unit capital and labour costs are $2
and $1, respectively. The firm is contracted to provide 4000 units of output and wants
to fulfil this contract at minimal cost.
(a) Explain briefly why the Lagrangian function is
g(K, L, l) = 2K + L + l[50 − KL]
(b) Write down expressions for
−g −g
−g
,
and .
−K −L
−l
(c) Find values of K and L which fulfil the quota with minimal cost to the firm.
5. A firm’s production function is given by
Q = 80KL
which is subject to a budgetary constraint, 3K + 5L = 1500.
➜
470 Chapter 5 Partial Differentiation
(a) Work out the first-order partial derivatives of the Lagrangian function
g(K, L, l) = 80KL + l[1500 − 3K − 5L]
(b) Find values of K and L which maximise output and satisfy the budgetary constraint.
(c) Estimate the increase in optimal output which results when the total budget of 1500
is increased to 1501.
6. A monopolistic producer of two goods, G1 and G2, has a total cost function
TC = 5Q1 + 10Q2
where Q1 and Q2 denote the quantities of G1 and G2, respectively. If P1 and P2 denote
the corresponding prices, then the demand equations are
P1 = 50 − Q1 − Q2
P2 = 100 − Q1 − 4Q2
Find the maximum profit if the firm’s total costs are fixed at $100. Estimate the new
optimal profit if total costs rise to $101.
Exercise 5.6*
1. (a) Use Lagrange multipliers to find the optimal value of
z = x + 2y
subject to the constraint
x + y2 = M
(b) Show that in this case the value of l gives the exact change in the optimal value of
z when M increases by 1 unit.
2. A production function is given by Q = 200K0.8L0.6.
(a) Find the optimal value of Q when the inputs are subject to a budgetary constraint
of 5K + 2L = 2000.
(b) Find the optimal value of input costs, C = 5K + 2L, when the inputs are constrained
by an output quota of Q = 8000.
3. (a) Find the maximum value of Q = K0.4L0.6 if the budgetary constraint is 3K + 5L = 3000.
(b) What happens to the optimal output if the available budget doubles?
4. A firm that manufactures speciality bicycles has a profit function
π = 5x2 − 10xy + 3y2 + 240x
where x denotes the number of frames and y denotes the number of wheels. Find the
maximum profit assuming that the firm does not want any spare frames or wheels left
over at the end of the production run.
Section 5.6 Lagrange multipliers 471
5. Find the maximum value of
Q 5 10 (KL)
subject to the cost constraint
K + 4L = 16
Estimate the change in the optimal value of Q if the cost constraint is changed to
K + 4L = 17
6. A consumer’s utility function is given by
U = a ln x1 + b ln x2
Find the values of x1 and x2 which maximise U subject to the budgetary constraint
P1x1 + P2x2 = M
7. An advertising agency spends $x on a newspaper campaign and a further $y promoting
its client’s products on local radio. It receives 15% commission on all sales that the
client receives. The agency has $10 000 to spend in total, and the client earns $M from
its sales, where
M5
100 000x
40 000y
1
50 1 x
30 1 y
Use the method of Lagrange multipliers to determine how much should be spent on
advertising in newspapers and on radio to maximise the agency’s net income. Give your
answers correct to two decimal places.
8. A firm produces three goods, A, B and C. The number of items of each are x, y and z,
respectively. The firm is committed to producing a total number of 30 items of types A
and B. The firm’s associated profit function is
π = 8x + 12y + 4z − 0.5x2 − 0.5y2 − z2
How many goods of each type must be produced to maximise profit subject to the
constraint?
9. A firm decides to invest x units of capital in project A and y units in project B. The
expected return for 1 unit of investment is $400 in project A and $800 in project B.
However, in order to meet the expectations of the firm’s ethical and environmental
policy, the values of x and y must satisfy the constraint
x2 + y2 − 4x − 6y = 67
How many units of each type should the firm buy in order to maximise total return?
10. Use the method of Lagrange multipliers to show that when a utility function, U = U(x1,x2),
is optimised subject to a budgetary constraint, P1x1 + P2x2 = M, the ratio of marginal
utility to price is the same for both goods.
11. A function, z = ax + by, is to be optimised subject to the constraint, x2 + y2 = 1 where
a and b are positive constants. Use Lagrange multipliers to show that this problem has
only one solution in the positive quadrant (i.e. in the region x > 0, y > 0) and that the
optimal value of z is a2 1 b2 .
Formal mathematics
In Section 5.1 we stated that the partial derivative, ∂f/∂x, is found by ‘differentiating f with
respect to x with y held constant’. Whilst this tells you exactly what to do to work it out, it is
very imprecise and gives no real geometrical insight into what it represents.
Look carefully at the surface, z = f(x, y) shown in Figure 5.11. Underneath the surface in
the (x, y)-plane is the line, y = b, and the point (a, b) which lies on this line. Vertically above
this line, etched on the surface itself, is a curve. As you move along this curve from F to G to
H, the y coordinate remains fixed at y = b and the height z varies, in accordance with any
changes in x. The gradient of the tangent to this curve at the point (a, b) measures the rate
of change of the function as x varies with y fixed. This is the value of the partial derivative
∂f /∂x evaluated at (a, b). If you imagine yourself walking on the surface along the line FH,
the gradient measures how steep the path is. In Figure 5.11 the curve is uphill at the point G
so fx(a, b) > 0. Also, the curve is concave, so the second-order partial derivative, fxx(a, b) < 0.
A similar interpretation can be given for the partial derivative with respect to y. It is the
gradient of the curve RS running parallel to the y axis in Figure 5.11.
We can use this geometrical interpretation to give a formal definition of a partial derivative.
In Figure 5.12 the point V is directly above (a, b) so is at a height, f(a, b). The point W is a
short distance away along the curve directly above (a + Δx, b), so it is at a height f(a + Δx, b).
The gradient of the chord VW is given by
f (a 1 Dx, b) 2 f (a, b)
Dx
z
S
G
F
y
R
b
a
H
(a, b)
x
Figure 5.11
Formal mathematics 473
z
V
W
y
b
a
a 1 Dx
x
Figure 5.12
As Δ x gets smaller and smaller, the gradient of the chord approaches that of the tangent so
we can define
fx (a, b) 5 lim
Dx
0
f (a 1 Dx, b) 2 f (a, b)
Dx
A similar argument gives
fy (a, b) 5 lim
Dy
0
f (a, b 1 Dy) 2 f (a, b)
Dy
Multiple choice questions
Question 1
If f(x, y) =2x7 + 3x2 y3, work out the value
A. 780
B. 340
−2f
at (2, 3).
−x−y
C. 126
D. 544
E. 324
Question 2
If the function f(x, y) = 4xy + x2y3 satisfies the equation 2x2
mine the value of n.
A. 2
B. 10
C. 6
2
−2 f
2− f
1 nxy 5 10f , deter2 1 y
−x
−y2
D. 8
E. 40
Question 3
Find the marginal rate of technical substitution for the Cobb–Douglas production function Q = AKa Lb.
A.
B.
C.
D.
E.
bK
aL
bL
aK
aK
bL
aL
bK
None of the above
Question 4
If the method of Lagrange multipliers is used to find the maximum value of 10xy subject
to the constraint 7x + 3y = 15, work out the value of the Lagrange multiplier.
A.
42
25
B.
25
7
C.
7
5
D.
3
7
E.
2
3
Question 5
Find the value of z at the minimum point of the function z = x3 + y3 − 24xy + 1000.
A. 978
B. 1000
C. 488
D. 1022
E. 968
Multiple choice questions 475
Question 6
The demand function of a good is given by Q = 100 − 6P + 0.1Y 2. Find the income elasticity of demand when P = 10 and Y = 20.
A. 2
B. 0.8
C. 0.5
D. 1
E. 0.75
Question 7
A consumer’s utility function is given by U = 4ln x1 + ln x2. Find the values of x1 and x2
which maximise U subject to the budgetary constraint x1 + 2x2 = 100.
A. x1 = 80, x2 = 10
B. x1 = 20, x2 = 40
C. x1 = 60, x2 = 20
D. x1 = 40, x2 = 30
E. x1 = 30, x2 = 70
Question 8
The supply and demand functions are given by
P = aQS + b
P = cQD + d
where a, b and d are positive constants, and c is a negative constant. Find the multiplier,
−Q
, for the equilibrium quantity.
−c
A.
21
a1c
B.
b2d
(a 2 c)2
C.
1
a1c
D.
1
(a 1 c)2
E.
d2b
(a 2 c)2
Question 9
The demand function of a good is given by Q = 100 − 3P + 2PA + 0.1Y where Q, P, PA
and Y denote quantity demanded, price of the good, price of an alternative good and
income, respectively. Find the cross-price elasticity of demand when P = 20, PA = 15 and
Y = 300.
A. 0.1
B. 0.2
C. 0.3
D. 0.4
E. 0.5
Question 10
The production function of a firm is given by Q = K 0.2 L0.7 where Q, K and L denote output,
capital and labour, respectively. The firm currently uses 300 units of capital and 500 units
of labour. If the firm decreases labour by 2.8, estimate the increase in capital needed to
maintain current output levels.
A. 5.4
B. 4.8
C. 6.4
D. 6.2
E. 5.9
476 Chapter 5 Partial Differentiation
Question 11
On an isoquant diagram, capital is plotted on the vertical axis and labour is plotted on the
horizontal axis. Unit prices of capital and labour are $24 and $18, respectively, and the firm has
a fixed amount to spend on these inputs. Work out the gradient of the associated isocost line.
A. −3/4
B. −4/3
C. −2/3
D. −3/2
E. Insufficient information
Question 12
An individual’s utility function for the consumption of two goods, 1 and 2, is given by
U 5 a x1 1 b x2 . Find an expression for the marginal rate of commodity substitution at
the point (4, 9).
A.
3a
4b
B.
a
2b
C.
5a
2b
D.
3a
2b
E.
5a
3b
Question 13
The price of the goods 1 and 2 in Question 12 are $1 and $2, respectively. If the individual has
a budget of $100 to spend on these goods, determine the optimal consumption of good 1.
A.
200a2
2a2 1 b2
B.
100a2
a2 1 2b2
C.
200a2
a2 1 2b2
D.
100a2
4a2 1 b2
E.
200a2
4a2 1 b2
Question 14
An increase in the price of a good by 5% leads to an increase of 2% in the demand for an
alternative good. Calculate the cross-price elasticity of demand and state whether these
goods are complementary or substitutable.
A. 2.5 and substitutable
B. −0.4 and complementary
C. 0.4 and substitutable
D. −2.5 and complementary
E. −0.4 and substitutable
Examination questions
Question 1
A function of two variables is given by f(x, y) = 2x + 3x2y − y2.
(a) Find expressions for the first-order and second-order partial derivatives of f.
(b) Evaluate fx(1, 2) and fy(1, 2), and use the small increments formula to estimate the
change in f as x increases by 0.2 and y decreases by 0.1 at the point (1, 2).
(c) A curve is defined implicitly by 2x + 3x2y − y2 = 4. Verify that (1, 2) lies on this curve
and find the gradient at this point.
Question 2
The demand function of a good is given by
Q = 1000 − 2P − 0.5PA2 + 0.01Y
where Q, P, PA and Y denote the quantity demanded, the price of the good, the price of an
alternative good and income, respectively.
Find the cross-price elasticity of demand and income elasticity of demand when P = 15,
PA = 20 and Y = 3000.
(a) Explain whether the good is inferior, normal or superior.
(b) Use the small increments formula to estimate the change in demand if P increases by
0.5, PA decreases by 0.25 and Y is unchanged.
Question 3
Sophie has $50 to spend on goods 1 and 2. The unit prices of these goods are $1 and $2,
respectively. Her utility function is given by U = x1x2 + 2x1 + 2x2
(a) Write down Sophie’s budgetary constraint and deduce that U = 48x2 − 2x22 + 100.
(b) Hence work out Sophie’s optimal consumption of these goods.
Question 4
1
4
A firm’s production function is given by Q 5 KL 1 ln(KL3).
(a) Write down expressions for the marginal product of labour and marginal product of
capital.
(b) The firm currently uses 18 units of capital and 50 units of labour, and decides to reduce
the level of labour to 49 units. Estimate the corresponding increase in capital if output
is to remain unchanged.
478 Chapter 5 Partial Differentiation
Question 5
A firm is the sole producer of two goods labelled 1 and 2. The demand functions are
given by
Q1 = 495 − P1 − P2
Q2 = 695 − P1 − 2P2
where Qi and Pi denote the quantity demanded and price of good i, respectively. The total
cost function is
TC = Q12 + Q1Q2 + Q22
(a) Are these goods substitutable or complementary?
(b) Show that the profit function is given by
π = 259Q1 + 200Q2 + Q1Q2 − 3Q21 − 2Q22
(c) Find the values of Q1 and Q2 which maximise the profit and verify that the second-
order conditions for a maximum are satisfied.
(d) Find the prices that the producer should charge to maximise profit.
Question 6
A firm’s unit capital and labour costs are $4 and $1, respectively. If the production function is given by Q 5 LK and total input costs are $120, use the method of Lagrange
multipliers to find the levels of K and L which maximise output.
Verify that the ratio of marginal products to price is the same for both inputs at the
optimum.
Question 7
A firm’s production function is given by Q 5 16 K 1 6 L where Q, K and L denote output, capital and labour, respectively. The cost of providing each unit of capital and labour
is $80 and $27, respectively.
Find the number of units of capital and labour if the firm wishes to minimise total costs
whilst satisfying a production quota of 102 units of output.
Question 8
Find and classify the stationary points of the function C(x, y) = (2xy − x2)e−y + 100.
Question 9
A firm’s production function is given by Q 5
labour, respectively.
K2L
where K and L denote capital and
K 1 4L
(a) Show that this production function is homogeneous of degree 2.
(b) Calculate the marginal products, MPK and MPL and hence verify that Euler’s theorem
holds for this function.
Examination questions 479
The firm currently uses 40 units of capital and 15 units of labour.
(c) Calculate the marginal rate of technical substitution and give a brief interpretation of
this number.
(d) The firm decides to reduce capital by 0.5 units. Use the small increments formula to
estimate the increase in labour needed to increase the total output by 1 unit.
Question 10
An individual needs to allocate the 168 hours available each week to work, W, hours and
free time, F, hours. He earns $20 an hour (after deductions), and his utility function is
given by U = 6E1/3 F2/3 where E denotes his weekly earnings.
Explain briefly why the Lagrangian for this problem may be written as
L(E, F, l) = 6E1/3 F2/3 + l(3360 − E − 20F)
Hence find the optimal number of hours worked in a week.
If the hourly rate of pay increases, does the optimal number of hours worked in a week
increase, decrease or stay the same?
Question 11
A firm intends to maximise the output Q which satisfies a production function of the form,
Q = f(K, L), subject to a budgetary constraint, 16K + 25L = 220 where K and L denote
capital and labour, respectively.
Work out the value of capital if L = 4 and the constraint is satisfied.
Calculate the slope of the constraint when K is plotted on the vertical axis and L is plotted on the horizontal axis.
Write down the marginal rate of technical substitution at the maximum and justify your
answer using a general diagram showing isoquants of the production function.
Question 12
(a) A function is defined implicitly by
x3 + 2xy + y4 = 61
Verify that the point (−2, 3) lies on this curve and find the gradient at this point.
(b) Find and classify the stationary point on the surface defined by
z = 4x2 − xy + 3y2 − 8x + y + 3
Question 13
An individual’s utility function is given by U 5
x1x2
where x1 and x2 denote the number
2x1 1 x2
of good 1 and the number of good 2 that she wishes to buy.
(a) Write down expressions for the marginal utilities,
−U
−U
and
, and show that the law
−x2
−x1
of diminishing marginal utilities holds for each of these goods.
480 Chapter 5 Partial Differentiation
(b) Calculate the marginal rate of commodity substitution for the indifference curve which
passes through (1, 3) and hence estimate the increase in x2 required to maintain the
current level of utility when x1 decreases by 0.02 units.
(c) Use Lagrange multipliers to find the optimal values of x1 and x2 if the individual maximises utility subject to a budgetary constraint, x1 + 8x2 = 100.
Question 14
A firm’s unit capital and labour costs are $1 and $2, respectively, and the total input costs
are fixed at $M. If the production function is given by Q = 8KL, use the method of
Lagrange multipliers to show that the optimum output is M2, and find the levels of K and
L at which it is achieved.
(a) Verify that the ratio of marginal product to price is the same for both inputs at the
optimum, and give an interpretation of this result.
(b) Verify that the Lagrange multiplier gives the approximate change in optimal output
when the input costs increase by one unit.
(c) Write down the equation of the isoquant that passes through the optimum point and
hence show that the cost constraint is a tangent to the isoquant at this point. Explain
why this was to be expected.
Question 15
(a) The equations for the IS model are given by:
Y=C+I
C = aY + b (0 < a < 1, b > 0)
I = cr + d (c < 0, d > 0)
where Y, C, I and r denote national income, consumption, investment and interest rate,
and a, b, c and d are parameters.
Show that when the commodity market is in equilibrium,
(1 − a)Y − cr = b + d
(b) The equations for the LM model are given by:
MS = MD
MS = M*S(M*S > 0)
(k1 > 0, k2 < 0, k3 > 0)
MD = k1Y + k2r + k3
where MS and MD denote the money supply and total demand for money.
Show that when the money market is in equilibrium,
k1Y + k2r = M*S − k3
(c) Use the results of parts (a) and (b) to show that when the commodity and money mar-
kets are both in equilibrium, national income is given by
Y5
k2(b 1 d ) 1 c(MS* 2 k3)
(1 2 a)k2 1 ck1
Write down an expression for the multiplier,
−Y
. Giving reasons for your answer, state
−b
whether national income increases or decreases as a result of an increase in autonomous
consumption.
This page intentionally left blank
Chapter 6
Integration
This chapter concludes the topic of calculus by considering the integration of functions of one
variable. It is in two sections, which should be read in the order that they appear.
Section 6.1 introduces the idea of integration as the opposite process to that of differentiation.
It enables you to recover an expression for the total revenue function from any given marginal
revenue function, to recover the total cost function from any marginal cost function and so on.
You will no doubt be pleased to discover that no new mathematical techniques are needed
for this. All that is required is for you to put your brain into reverse gear. Of course, driving
backwards is a little harder to master than going forwards. However, with practice you should
find integration almost as easy as differentiation.
Section 6.2 shows how integration can be used to find the area under the graph of a function.
This process is called definite integration. We can apply the technique to supply and demand
curves and so calculate producer’s and consumer’s surpluses. Definite integration can also be
used to determine capital stock and to discount a continuous revenue stream.
Section 6.1
Indefinite integration
Objectives
At the end of this section you should be able to:
●
Recognise the notation for indefinite integration.
●
Write down the integrals of simple power and exponential functions.
●
Integrate functions of the form af(x) + bg(x).
●
Find the total cost function given any marginal cost function.
●
Find the total revenue function given any marginal revenue function.
●
Find the consumption and savings functions given either the marginal propensity to consume or
the marginal propensity to save.
●
Use the method of inspection to integrate more complicated functions.
Throughout mathematics there are many pairs of operations which cancel each other out and
take you back to where you started. Perhaps the most obvious pair is multiplication and division.
If you multiply a number by a non-zero constant, k, and then divide by k, you end up with
the number you first thought of. This situation is described by saying that the two operations
are inverses of each other. In calculus, the inverse of differentiation is called integration.
Suppose that you are required to find a function, F(x), which differentiates to
f(x) = 3x 2
Can you guess what F(x) is in this case? Given such a simple function, it is straightforward
to write down the answer by inspection. It is
F(x) = x 3
because
F′(x) = 3x 2 = f(x) ✓
as required.
As a second example, consider
f(x) = x7
Can you think of a function, F(x), which differentiates to this? Recall that when power functions are differentiated, the power decreases by 1, so it makes sense to do the opposite here
and to try
F(x) = x 8
Unfortunately, this does not quite work out, because it differentiates to
8x7
which is eight times too big. This suggests that we try
F(x) = 1/8 x 8
Section 6.1 Indefinite integration 485
which does work because
F′(x) = 8/8 x7 = x7 = f(x) ✓
In general, if F′(x) = f(x), then F(x), is said to be the integral (sometimes called the anti-derivative
or primitive) of f(x) and is written
F(x) =
 f(x)dx
read ‘integral of f
of x dee x’
In this notation
3x dx = x
2
3
and
x dx = / x
7
1
8
8
Here is a problem for you to try. Do not let the notation
dx
put you off. It is merely an instruction for you to think of a function that differentiates to
whatever is squashed between the integral sign ‘∫’ and dx. If you get stuck, try adding 1 on
to the power. Differentiate your guess, and if it does not quite work out, then go back and
try again, adjusting the coefficient accordingly.
Practice Problem
1. Find





(a) 2xdx (b) 4x3dx (c) 100x99dx (d) x3dx (e) x18dx
In Practice Problem 1(a) you probably wrote
2xdx = x
2
However, there are other possibilities. For example, both of the functions
x 2 + 6 and x 2 − 59
differentiate to 2x, because constants differentiate to zero. In fact, we can add any constant,
c, to x2 to obtain a function that differentiates to 2x. Hence
2xdx = x + c
2
486 Chapter 6 Integration
The arbitrary constant, c, is called the constant of integration. In general, if F(x) is any function that differentiates to f(x), then so does
F(x) + c
Hence
f (x)dx 5 F (x) 1 c
In Practice Problem 1 you used guesswork to find various integrals. In theory most integrals
can be worked out in this way. However, considerable ingenuity (and luck!) may be required
when integrating complicated functions. It is possible to develop various rules similar to
those of differentiation, which we discussed in Chapter 4, although even then we sometimes
have to resort to sheer trickery. It is not our intention to plod through each rule as we did in
Chapter 4, for the simple reason that they are rarely needed in economics. However, it is
worthwhile showing you a direct way of integrating simple functions such as
2x − 3x 2 + 10x 3 and x − e2x + 5
We begin by finding general formulae for
 x dx
n
and
e
mx
dx
To integrate f(x) = x n an obvious first guess is
F(x) = x n+1
This gives
F′(x) = (n + 1)x n
which is n + 1 times too big. This suggests that we try again with
F(x) 5
1
x n1 1
n11
which checks out because
F9(x) 5
n11 n
x 5 x n 5 f (x) ✓
n11
Hence
x ndx 5
1
x n11 1 c
n 11
To integrate a power function you simply add 1 to the power and divide by the number
you get. This formula holds whenever n is positive, negative, a whole number or a fraction.
There is just one exception to the rule, when n = −1. The formula cannot be used to integrate
1
x
because it is impossible to divide by zero. An alternative result is therefore required in this
case. We know from Chapter 4 that the natural logarithm function
ln x
Section 6.1 Indefinite integration 487
differentiates to give
1
x
and so
1
dx 5 ln x 1 c
x
The last basic integral that we wish to determine is
e
mx
dx
In Section 4.8 we showed that to differentiate an exponential function all we need to do is to
multiply by the coefficient of x. To integrate we do exactly the opposite and divide by the
coefficient of x, so
emxdx 5
1 mx
e 1c
m
It is easy to check that this is correct, because if
F(x) 5
1 mx
e
m
then
F9(x) 5
m mx
e 5 emx ✓
m
Example
Find

(a) x6dx (b)
# x dx (c) #
1
2

x dx (d) e2xdx
Solution
The formula
#x dx 5 n 1 1 x
1
n
n11
1c
can be used to find the first three integrals by substituting particular values for n.
(a) Putting n = 6 gives
#x dx 5 7 x 1 c
6
1
7
➜
488 Chapter 6 Integration
(b) Putting n = −2 gives
# x dx 5 #x
1
22
2
dx 5
1 21
1
x 1c52 1c
21
x
(c) Putting n = 1/2 gives
#
#
xdx 5 x 1/2 dx 5
2
1 3/2
x 1 c 5 x3/2 1 c
3/ 2
3
(d) To find
#e
2x
dx
we put m = 2 into the formula
#e
mx
dx 5
1 mx
e 1c
m
to get
#e dx 5 2 e
2x
1
2x
1c
Practice Problem
2. Find

# x dx (c) x
1
(a) x4dx (b)

(e) 1dx
1/3

(f) xdx

dx (d) e3xdx
3
# x dx
(g)
1
[Hint: in parts (b), (e) and (f ) note that 1/x3 = x−3, 1 = x0 and x = x1, respectively.]
In Section 4.2 we described three rules of differentiation known as the constant, sum and
difference rules. Given that integration is the inverse operation, these three rules also apply
whenever we integrate a function. The integral of a constant multiple of a function is
obtained by integrating the function and multiplying by the constant. The integral of the sum
(or difference) of two functions is obtained by integrating the functions separately and adding
(or subtracting). These three rules can be combined into the single rule:
[af(x) 1 bg(x)]dx 5 a f(x)dx 1 b g(x)dx
This enables us to integrate an expression ‘term by term’, as the following example
demonstrates.
Section 6.1 Indefinite integration 489
Example
Find
#
(a) (2x 2 2 4x6)dx (b)
# 7e
2x
1
#
2
2
dx (c) (5x 1 3x 1 2)dx
x
Solution
#
#
#
(a) (2x 2 2 4x 6)dx 5 2 x 2 dx 2 4 x 6dx
Putting n = 2 and n = 6 into
#x dx 5 n 1 1 x
1
n
n1 1
#x dx 5 3 x and #x dx 5 7 x
gives
1
2
3
1
6
7
Hence
#(2x 2 4x )dx 5 3 x 2 7 x
2
6
2
3
4
7
Finally, we add an arbitrary constant to get
#(2x 2 4x )dx 5 3 x 2 7 x 1 c
2
6
2
3
4
7
As a check:
if F(x) 5
(b)
# 7e
2x
1
2 3 4 7
x 2 x 1 c then F9(x) 5 2x 2 2 4x 6 ✓
3
7
#
#
2
1
dx
dx 5 7 e2xdx 1 2
x
x
Now
#e
mx
dx 5
1 mx
e
m
so putting m = −1 gives
#e
2x
dx 5
1 2x
e 5 2e2x
21
Also, we know that the reciprocal function integrates to the natural logarithm function, so
# x dx 5 ln x
1
Hence
# 7e
2x
1
2
dx 5 27e2x 1 2 ln x
x
➜
490 Chapter 6 Integration
Finally, we add an arbitrary constant to get
# 7e
2x
1
2
dx 5 27e2x 1 2 ln x 1 c
x
As a check:
2
x
if F(x) 5 27e2x 1 2 ln x 1 c then F9(x) 5 7e2x 1 ✓
#
#
#
#
(c) (5x 2 1 3x 1 2)dx 5 5 x 2 dx 1 3 xdx 1 2 1dx
Putting n = 2, 1 and 0 into
#x dx 5 n 1 1 x
1
n
n11
gives
#x dx 5 3 x , #xdx 5 2 x , #1dx 5 x
1
2
3
1
2
Hence
#(5x 1 3x 1 2)dx 5 3 x 1 2 x 1 2x
2
5
3
3
2
Finally, we add an arbitrary constant to get
#(5x 1 3x 1 2)dx 5 3 x 1 2 x 1 2x 1 c
2
5
3
3
2
As a check:
if F(x) 5
5 3 3 2
x 1 x 1 2x 1 c then F9(x) 5 5x 2 1 3x 1 2 ✓
3
2
Advice
We have written out the solution to this example in detail to show you exactly how integration is
performed. With practice you will probably find that you can just write the answer down in a single
line of working, although it is always a good idea to check (at least in your head, if not on paper), by
differentiating your answer, that you have not made any mistakes.
The technique of integration that we have investigated produces a function of x. In the
next section a different type of integration is discussed which produces a single number as
the end result. For this reason we use the word indefinite to describe the type of integration
considered here to distinguish it from the definite integration in Section 6.2.
Section 6.1 Indefinite integration 491
Practice Problem
3. Find the indefinite integrals
#
(a) (2x 2 4x 3)dx (b)
# 10x 1 x5
4
2
#
dx (c) (7x 2 2 3x 1 2)dx
In Section 4.3 we described several applications of differentiation to economics. Starting
with any basic economic function, we can differentiate to obtain the corresponding marginal
function. Integration allows us to work backwards and to recover the original function from
any marginal function. For example, by integrating a marginal cost function, the total cost
function is found. Likewise, given a marginal revenue function, integration enables us to
determine the total revenue function, which in turn can be used to find the demand function.
These ideas are illustrated in the following example, which also shows how the constant of
integration can be given a specific numerical value in economic problems.
Example
(a) A firm’s marginal cost function is
MC = Q2 + 2Q + 4
Find the total cost function if the fixed costs are 100.
(b) The marginal revenue function of a monopolistic producer is
MR = 10 − 4Q
Find the total revenue function and deduce the corresponding demand function.
(c) Find an expression for the consumption function if the marginal propensity to con-
sume is given by
MPC 5 0.5 1
0.1
Y
and consumption is 85 when income is 100.
Solution
(a) We need to find the total cost from the marginal cost function
MC = Q2 + 2Q + 4
Now
MC 5
d(TC)
dQ
so
#
#
TC 5 MCdQ 5 (Q 2 1 2Q 1 4)dQ 5
Q3
1 Q 2 1 4Q 1 c
3
➜
492 Chapter 6 Integration
The fixed costs are given to be 100. These are independent of the number of goods
produced and represent the costs incurred when the firm does not produce any goods
whatsoever. Putting Q = 0 into the TC function gives
03
1 02 1 4(0) 1 c 5 c
3
TC 5
The constant of integration is therefore equal to the fixed costs of production, so c = 100.
Hence
Q3
1 Q 2 1 4Q 1 100
3
TC 5
(b) We need to find the total revenue from the marginal revenue function
MR = 10 − 4Q
Now
MR 5
d(TR)
dQ
so
#
#
TR 5 MRdQ 5 (10 2 4Q)dQ 5 10Q 2 2Q 2 1 c
Unlike in part (a) of this example, we have not been given any additional information
to help us to pin down the value of c. We do know, however, that when the firm produces
no goods, the revenue is zero, so that TR = 0 when Q = 0. Putting this condition into
TR = 10Q − 2Q2 + c
gives
0 = 10(0) − 2(0)2 + c = c
The constant of integration is therefore equal to zero. Hence
TR = 10Q − 2Q2
Finally, we can deduce the demand equation from this. To find an expression for total
revenue from any given demand equation, we normally multiply by Q, because
TR = PQ. This time we work backwards, so we divide by Q to get
P5
TR 10Q 2 2Q2
5 10 2 2Q
5
Q
Q
so the demand function is
P = 10 − 2Q
(c) We need to find consumption given that the marginal propensity to consume is
MPC 5 0.5 1
0.1
Y
Now
MPC 5
dC
dY
so
#
C 5 MPCdY 5
# 0.5 1 0.1Y
dY 5 0.5Y 1 0.2 Y 1 c
Section 6.1 Indefinite integration 493
where the second term is found from
# 0.1Y dY 5 0.1#Y
21/2
dY 5 0.1
1 1/2
Y
5 0.2 Y
1/2
The constant of integration can be calculated from the additional information that
C = 85 when Y = 100. Putting Y = 100 into the expression for C gives
85 5 0.5(100) 1 0.2 100 1 c 5 52 1 c
and so
c = 85 − 52 = 33
Hence
C 5 0.5Y 1 0.2 Y 1 33
Practice Problem
4. (a) A firm’s marginal cost function is
MC = 2
Find an expression for the total cost function if the fixed costs are 500. Hence find
the total cost of producing 40 goods.
(b) The marginal revenue function of a monopolistic producer is
MR = 100 − 6Q
Find the total revenue function and deduce the corresponding demand function.
(c) Find an expression for the savings function if the marginal propensity to save is
given by
MPS = 0.4 − 0.1Y−1/2
and savings are zero when income is 100.
So far, we have seen how easy it is to integrate a function written as a linear combination
of power functions. We began the chapter by using ‘guesswork’ to integrate the simple power
function x7. We did this by trying x8 and then fiddled around with the numerical coefficient
to make it work. The same approach can be used to integrate a wide variety of more complicated functions. To find
(2x + 7) dx
5
we would increase the power by 1 and so try (2x + 7)6 in the first instance. This differentiates to
6(2x + 7)5 × 2 = 12(2x + 7)5 (by the chain rule)
This is 12 times too big, so we go back to the original guess and divide it by 12 to get
#(2x 1 7)
5
dx 5
1
s2x 1 7d6 1 c
12
This approach is variously known as the method of ‘guesswork’, ‘inspection’ and ‘recognition’.
As a second example, consider the task of trying to think of a function which differentiates to
x2(1 + x3)4
494 Chapter 6 Integration
At first glance this appears to be fiendishly difficult, but help is at hand from the chain rule.
Notice that the term outside the brackets is more or less the same as the derivative of the term
inside the brackets. The inside, 1 + x3, actually differentiates to 3x2 which, apart from the 3,
agrees with the factor in front of the brackets. We are now in a position to make a guess for
the integral of x2(1 + x3)4. We can mentally ignore the x2 (knowing that when we get round to
differentiating our guess, it will be automatically generated from the chain rule) and concentrate on integrating the brackets. This is a power of 4, so it makes sense to try differentiating
the same bracket raised to a power of 5. Our first guess for the integral is therefore
F(x) = (1 + x3)5
and when we use the chain rule to differentiate this, we get
F′(x) = 5(1 + x3)4 × 3x2 = 15x2 (1 + x3)4
This is nearly right but is 15 times too big, so it suggests that we try
F (x) 5
1
1 1 x3
15
5
which does work because
F'(x) 5
1
3 5 1 1 x3
15
4
3 3x2 5 x 2 1 1 x3
4
Hence
# x (1 1 x ) dx 5 151 1 1 x
2
3 5
3 4
1c
Most of what we have written down above can be ticked off in your head and the answer just
written straight down.
Example
Find
(a)

# 114x
x
3
x2e−x dx (b)
2 3
dx (c)
#
x9
dx
11x10
Solution
(a) The exponent −x3 basically differentiates to x2 (apart from an extra numerical factor,
−3, which can be sorted out later). The chain rule will generate this term so can just
concentrate on the exponential function which integrates to itself. Our first guess is
3
F(x) = e−x
which differentiates to give
3
F′(x) = −3x2e−x
This is −3 times too big, so
#x e
2 2x3
1 3
dx 5 2 e2x 1 c
3
(b) The inside of the brackets, 1 + 4x2, differentiates to 8x which (apart from the numerical
coefficient) coincides with the top of the fraction. We can therefore just concentrate on
the rest of the function which is (1 + 4x2)−3. An obvious guess is obtained by increasing
the power by 1, so we try
F(x) = (1 + 4x2)−2
Section 6.1 Indefinite integration 495
which differentiates to give
F9(x) 5 22 1 1 4x2
23
3 8x 5
216x
1 1 4x2
3
This is −16 times too big, so
# 1 1x4x
2 3
dx 5
21
16 1 1 4x2
2
1c
(c) For the integral,
#
x9
dx
1 1 x10
we notice that the top of the fraction is basically the derivative of the bottom. Reciprocal
functions like this integrate to natural logarithms, so our first guess at the answer is
F(x) = ln(1 + x10)
The chain rule automatically gives us the extra x9 that we need. Differentiating our
guess gives
F9(x) 5
1
10x9
9
10 3 10x 5
11x
1 1 x10
This is 10 times too big, so
#
x9
1
ln(1 1 x10) 1 c
dx 5
10
1 1 x10
Practice Problem
5. Find
(a)

(5x + 1)3dx (b)

x(1 + x2)7dx (c)
#
4x3
dx (d)
2 1 x4
e (1 + e ) dx
x
x 3
Key Terms
Anti-derivative A function whose derivative is a given function.
Constant of integration
indefinite integral.
The arbitrary constant that appears in an expression when finding an
Definite integration The process of finding the area under a graph by subtracting the values
obtained when the limits are substituted into the anti-derivative.
Indefinite integration
Integral
The number
Integration
The process of obtaining an anti-derivative.
b
a
f (x)dx (definite integral) or the function f(x)dx (indefinite integral).
The generic name for the evaluation of definite or indefinite integrals.
Inverse (operation) The operation that reverses the effect of a given operation and takes you
back to the original. For example, the inverse of halving is doubling.
Primitive An alternative word for an anti-derivative.
496 Chapter 6 Integration
Exercise 6.1
1. Find
#
(b) x 4 dx
(d)
#
# 1x dx
(e) x 3/ 2 dx
#
(h) (ax 1 b)dx
#
(a) 6 x 5dx
(c) 10e10 x dx
#
(g) (x2 2 8x 1 3)dx
#
(f) (2x 3 2 6x)dx
#
(i)
# 7x 1 4e
3
22x
2
3
dx
x2
2. (a) Find the total cost if the marginal cost is
MC = Q + 5
and fixed costs are 20.
(b) Find the total cost if the marginal cost is
MC = 3e0.5Q
and fixed costs are 10.
3. The marginal cost function is given by
MC = 2Q + 6
If the total cost is 212 when Q = 8, find the total cost when Q = 14.
4. Find the total revenue and demand functions corresponding to each of the following
marginal revenue functions:
(a) MR = 20 − 2Q
(b) MR 5
6
Q
5. Find the consumption function if the marginal propensity to consume is 0.6 and
consumption is 10 when income is 5. Deduce the corresponding savings function.
6. Find the short-run production functions corresponding to each of the following
marginal product of labour functions:
(a) 1000 − 3L2
(b)
6
L
2 0.01
7. A firm’s marginal revenue and marginal cost functions are given by
MR = 10 − 4Q and MC = 1
If fixed costs are 4, find the profit when Q = 2.
8. (1) Differentiate
F(x) = (2x + 1)5
Hence find
(2x + 1) dx
4
Section 6.1 Indefinite integration 497
(2) Use the approach suggested in part (1) to find
#
#
(a) (3x 2 2)7 dx (b) (2 2 4x)9 dx (c)
#
2x 1 1dx (d)
# 7x 11 3 dx
Exercise 6.1*
1. Find
#
#
#
(a) x(x 5 2 2)dx (b) x10 2 3 x 1 e2x dx (c) x 3 2
5
2
1 2 4e24x dx
x6
x
2. (a) Find the consumption function given that
MPC 5 20 1
10
Y 3/4
and that consumption is 420 when Y = 16.
(b) If the marginal cost is
MC = 15 + 3Q2
find an expression for the variable cost per unit.
3. Find the total cost function, TC, when the fixed costs are C and the marginal cost is
(a) aQ + b (b) aebQ
4. Use integration by inspection to write down the indefinite integral of:
(a)
(4x − 7)5 (b)
x3
(e)
1 1 x4
2
1
(c)
3x 1 1
2
(f) xe−x (g)
1
(d) x(1 + x2)4
2x 1 3
2
1 1 lnx
(h)
x
x2
x3 1 5
5. (a) Show that
x ( x 1x 2) 5 x 1 x 5/2
Hence find
#
x ( x 1 x 2)dx
(b) Use the approach suggested in part (a) to integrate each of the following functions:
x4 x6 1
1
, e2x(e3x 1 e2x 1 3),
x2
x 3/2
x2
1
x
6. (a) Show that
x4 2 x2 1 x
5 x3 2 x 1 x21/2
x
➜
498 Chapter 6 Integration
Hence find
#
x4 2 x2 1 x
dx
x
(b) Use the approach suggested in part (a) to integrate each of the following functions:
x2 2 x
,
x3
ex 2 e2x
,
e2x
x 2 x x 1 x2
x x
7. Find an expression for the savings function when
MPC 5 0.4 1
0.4
Y
and C 5 50 when Y 5 100
8. Find f(x) when f ″(x) = 6x, f(0) = 2 and f ′(0) = −4.
9. (a) Differentiate the function, x ln x.
(b) Use the result of part (a) to find ∫ ln x dx.
10. The number of units, N, produced each hour by a new employee after t hours on the
job satisfies
dN
5 10e20.1t
dt
Assuming that she is unable to produce any goods at the beginning of her shift, calculate how many she can produce per hour after eight hours. How many units per hour
will she be able to produce in the long run?
11. The rate of growth in the number of shops, n, in a chain of local grocery stores after t
months satisfies
3
dn
5
dt
t
Initially there are two stores.
Estimate the number of shops in the chain after 9 months.
12. A firm’s marginal revenue and marginal cost functions are given by
MR = 240 − 0.6Q2
and
MC = 150 + 0.3Q2
If fixed costs are 50, determine the maximum profit.
13. Use integration by inspection to find:
(a)
(ax + b) dx
(n ≠ −1) (b)
n
14. Use differentiation to show that
e
ax + b
# 2x41x 3 dx 5 ln(2x 1 3) 1 2x 31 3 1 c
(b) x e dx = e (x − 2x + 2) + c
(a)
2
2 x
x
2
dx (c)
# ax 11 b dx
Section 6.2
Definite integration
Objectives
At the end of this section you should be able to:
●
Recognise the notation for definite integration.
●
Evaluate definite integrals in simple cases.
●
Calculate the consumer’s surplus.
●
Calculate the producer’s surplus.
●
Calculate the capital stock formation.
●
Calculate the present value of a continuous revenue stream.
One rather tedious task that you may remember from school is that of finding areas. Sketched
in Figure 6.1 is a region bounded by the curve y = x2, the lines x = 1 and x = 2, and the x axis.
At school you may well have been asked to find the area of this region by ‘counting’ squares
on graph paper. A much quicker and more accurate way of calculating this area is to use
inte­g­ration. We begin by integrating the function
f(x) = x 2
y
y 5 x2
4
3
2
1
1
Figure 6.1
2
x
500 Chapter 6 Integration
to get
F(x) 5
1 3
x
3
In our case we want to find the area under the curve between x = 1 and x = 2, so we evaluate
1 3
(1) 5
3
1
F(2) 5 (2)3 5
3
F(1) 5
1
3
8
3
Finally, we subtract F(1) from F(2) to get
F(2) 2 F(1) 5
8
1
7
2 5
3
3
3
This number is the exact value of the area of the region sketched in Figure 6.1. Given the
connection with integration, we write this area as
2
# x dx
2
1
In general, the definite integral
b
# f(x)dx
a
denotes the area under the graph of f (x) between x = a and x = b as shown in Figure 6.2. The
numbers a and b are called the limits of integration, and it is assumed throughout this section
that a < b and that f (x) ≥ 0 as indicated in Figure 6.2.
The technique of evaluating definite integrals is as follows. A function F(x) is found which
differentiates to f(x). Methods of obtaining F(x) have already been described in Section 6.1.
The new function, F(x), is then evaluated at the limits x = a and x = b to get F(a) and F(b).
Finally, the second number is subtracted from the first to get the answer
F(b) − F(a)
In symbols,
b
# f (x)dx 5 F(b) 2 F(a)
a
y
y 5 f(x)
a
Figure 6.2
b
x
Section 6.2 Definite integration 501
The process of evaluating a function at two distinct values of x and subtracting one from
the other occurs sufficiently frequently in mathematics to warrant a special notation. We write
[F(x)]ba
as an abbreviation for F(b) − F(a), so that definite integrals are evaluated as
b
# f (x)dx 5 [F(x)] 5 F(b) 2 F(a)
b
a
a
where F(x) is the indefinite integral of f (x). Using this notation, the evaluation of
2
# x dx
2
1
would be written as
#
2
1 3
x
x dx 5
3
2
2
1
5
1
1 3 1 3 7
(2) 2 (1) 5
3
3
3
Note that it is not necessary to include the constant of integration, because it cancels out
when we subtract F(a) from F(b).
Example
Evaluate the definite integrals
#
# (x 1 1)dx
2
6
(a)
(b)
3dx
0
2
Solution
# 3dx 5 [3x] 5 3(6) 2 3(2) 5 12
6
(a)
6
2
2
This value can be confirmed graphically. Figure 6.3 shows the region under the graph
of y = 3 between x = 2 and x = 6. This is a rectangle, so its area can be found from the
formula
area = base × height
which gives
area = 4 × 3 = 12 ✓
y
y53
3
3
2
Figure 6.3
4
6
x
➜
502 Chapter 6 Integration
#
2
(b)
(x 1 1)dx 5
0
x2
1x
2
2
5
0
02
22
10 54
12 2
2
2
Again, this value can be confirmed graphically. Figure 6.4(a) shows the region under
the graph of y = x + 1 between x = 0 and x = 2. This can also be regarded as one-half of
the rectangle illustrated in Figure 6.4(b). This rectangle has a base of 2 units and a
height of 4 units, so it has area
2×4=8
The area of the region shown in Figure 6.4(a) is therefore
1
/2 × 8 = 4 ✓
(a)
(b)
y
1
y5 x11
3
3
3
1
1
2
x
2
Figure 6.4
In this example we deliberately chose two very simple functions so that we could demonstrate the fact that definite integrals really do give the areas under graphs. The beauty of the
integration technique, however, is that it can be used to calculate areas under quite complicated functions for which alternative methods would fail to produce the exact value.
Practice Problem
1. Evaluate the following definite integrals:
1
(a)
#
x3dx (b)
0
#
5
#
1
4
(2x 1 1)dx (c)
2
1
(d)
(x 2 2 x 1 1)dx
# e dx
x
0
To illustrate the applicability of definite integration, we concentrate on four topics:
●
consumer’s surplus;
●
producer’s surplus;
●
investment flow;
●
discounting.
We consider each of these in turn.
Section 6.2 Definite integration 503
6.2.1 Consumer’s surplus
The demand function, P = f (Q), sketched in Figure 6.5, gives the different prices that con­
sumers are prepared to pay for various quantities of a good. At Q = Q0 the price P = P0. The
total amount of money spent on Q0 goods is then Q0P0, which is given by the area of the
rectangle OABC. Now, P0 is the price that consumers are prepared to pay for the last unit
that they buy, which is the Q0th good. For quantities up to Q0 they would actually be willing
to pay the higher price given by the demand curve. The shaded area BCD therefore represents
the benefit to the consumer of paying the fixed price of P0 and is called the consumer’s surplus,
CS. The value of CS can be found by observing that
area BCD = area OABD − area OABC
P
D
P0
C
B
O
A
demand
curve
Q0
Q
Figure 6.5
The area OABD is the area under the demand curve P = f(Q), between Q = 0 and Q = Q0,
and so is equal to
Q0
#
f (Q)dQ
0
The region OABC is a rectangle with base Q0 and height P0 so
area OABC = Q0 P0
Hence
Q0
CS 5
#
f (Q)dQ 2 Q0 P0
0
Example
Find the consumer’s surplus at Q = 5 for the demand function
P = 30 − 4Q
➜
504 Chapter 6 Integration
Solution
In this case
f (Q) = 30 − 4Q
and Q0 = 5. The price is easily found by substituting Q = 5 into
P = 30 − 4Q
to get
P0 = 30 − 4(5) = 10
The formula for consumer’s surplus
Q0
CS 5
#
f (Q)dQ 2 Q0 P0
0
gives
CS 5
#
5
(30 2 4Q)dQ 2 5(10) 5 [30Q 2 2Q2]05 2 50
0
5 [30(5) 2 2(5)2] 2 [30(0) 2 2(0)2] 2 50
5 50
Practice Problem
2. Find the consumer’s surplus at Q = 8 for the demand function
P = 100 − Q2
6.2.2 Producer’s surplus
The supply function, P = g(Q), sketched in Figure 6.6 gives the different prices at which pro­
ducers are prepared to supply various quantities of a good. At Q = Q0 the price P = P0.
Assuming that all goods are sold, the total amount of money received is then Q0 P0, which is
given by the area of the rectangle OABC.
Now, P0 is the price at which the producer is prepared to supply the last unit, which is the
Q0th good. For quantities up to Q0 they would actually be willing to accept the lower price
given by the supply curve. The shaded area BCD therefore represents the benefit to the producer of selling at the fixed price of P0 and is called the producer’s surplus, PS. The value of PS
is found by observing that
area BCD = area OABC − area OABD
The region OABC is a rectangle with base Q0 and height P0, so
area OABC = Q0 P0
The area OABD is the area under the supply curve P = g(Q), between Q = 0 and Q = Q0, and
so is equal to
Section 6.2 Definite integration 505
P
P0
C
B
supply
curve
D
O
A
Q0
Q
Figure 6.6
Q0
#
g(Q)dQ
0
Hence
Q
PS 5 Q0P0 2
0
g(Q)dQ
0
Example
Given the demand function
P = 35 − QD2
and supply function
P = 3 + Q2S
find the producer’s surplus assuming pure competition.
Solution
On the assumption of pure competition, the price is determined by the market. Therefore,
before we can calculate the producer’s surplus, we need to find the market equilibrium
price and quantity. Denoting the common value of QD and QS by Q, the demand and supply functions are
P = 35 − Q2
and
P = 3 + Q2
so that
35 − Q2 = 3 + Q2
35 − 2Q = 3
2
−2Q2 = −32
Q = 16
2
(both sides are equal to P)
(subtract Q2 from both sides)
(subtract 35 from both sides)
(divide both sides by −2)
➜
506 Chapter 6 Integration
which has solution Q = ±4. We can obviously ignore the negative solution because it does
not make economic sense. The equilibrium quantity is therefore equal to 4. The corres­
ponding price can be found by substituting this into either the demand or the supply
equation. From the demand equation we have
P0 = 35 − (4)2 = 19
The formula for the producer’s surplus,
Q0
PS 5 Q0P0 2
#
g(Q)dQ
0
gives
# (3 1 Q )dQ
4
PS 5 4(19) 2
2
0
5 76 2 3Q 1
Q3
3
4
0
5 76 2 {[3(4) 1 1/3(4)3] 2 [3(0) 2 1/3(0)3]}
5 422/3
Practice Problem
3. Given the demand equation
P = 50 − 2QD
and supply equation
P = 10 + 2QS
calculate
(a) the consumer’s surplus (b) the producer’s surplus
assuming pure competition.
6.2.3 Investment flow
Net investment, I, is defined to be the rate of change of capital stock, K, so that
I5
dK
dt
Here, I(t) denotes the flow of money, measured in dollars per year, and K(t) is the amount of
capital accumulated at time t as a result of this investment flow and is measured in dollars.
Given a formula for capital stock in terms of time, we simply differentiate to find net
investment. Conversely, if we know the net investment function, then we integrate to find
the capital stock. In particular, to calculate the capital formation during the time period from
t = t1 to t = t2, we evaluate the definite integral
t2
# I(t)dt
t1
Section 6.2 Definite integration 507
Example
If the investment flow is
I(t) 5 9000 t
calculate
(a) the capital formation from the end of the first year to the end of the fourth year;
(b) the number of years required before the capital stock exceeds $100 000.
Solution
(a) In this part we need to calculate the capital formation from t = 1 to t = 4, so we evaluate
the definite integral
#
4
9000 tdt 5 9000
1
#
4
t 1/2dt 5 9000
1
2 3/2
t
3
4
5 9000
1
16
2
5 $42 000
2
3
3
(b) In this part we need to calculate the number of years required to accumulate a total of
$100 000. After T years the capital stock is
T
T
# 9000 tdt 5 9000 # t
1/2
0
dt
0
We want to find the value of T so that
#t
T
9000
1/2
dt 5 100 000
0
The integral is easily evaluated as
9000
2 3/2
t
3
T
5 9000
0
2 3/2 2 3/2
5 6000T 3/2
T 2 (0)
3
3
so T satisfies
6000T3/2 = 100 000
This non-linear equation can be solved by dividing both sides by 6000 to get
T3/2 = 16.67
and then raising both sides to the power of 2/3, which gives
T = 6.5
The capital stock reaches the $100 000 level about halfway through the seventh year.
Practice Problem
4. If the net investment function is given by
I(t) = 800t1/3
calculate
(a) the capital formation from the end of the first year to the end of the eighth year;
(b) the number of years required before the capital stock exceeds $48 600.
508 Chapter 6 Integration
6.2.4 Discounting
In Chapter 3 the formula
P = Se−rt/100
was used to calculate the present value, P, when a single future value, S, is discounted at r%
interest continuously for t years. We also discussed the idea of an annuity. This is a fund that
provides a series of discrete regular payments, and we showed how to calculate the original
lump sum needed to secure these payments for a prescribed number of years. This amount is
called the present value of the annuity. If the fund is to provide a continuous revenue stream
for n years at an annual rate of S dollars per year, then the present value can be found by
evaluating the definite integral
n
P5
# Se
2rt/100
dt
0
For example, if the discount rate is 9%, the present value of a continuous revenue stream
of a $1000 a year for five years is
# 1000e
5
P5
20.09t
dt
0
#e
5
5 1000
20.09t
dt
0
5 1000
21 20.09t
e
0.09
52
1000 20.09t 5
]0
[e
0.09
52
1000 20.45
2 1)
(e
0.09
5
0
e0 5 1
5 $4026.35
Practice Problem
5. Calculate the present value of a continuous revenue stream for 10 years at a constant
rate of $5000 per year if the discount rate is 6%.
Section 6.2 Definite integration 509
Key Terms
Consumer’s surplus The excess cost that a person would have been prepared to pay for goods
over and above what is actually paid.
b
Definite integral The number
between x = a and x = b.
The numbers a and b which appear in the definite integral,
Limits of integration
Net investment
f(x)dx which represents the area under the graph of f(x)
a
Rate of change of capital stock over time: I = dK/dt.
b
a
f(x)dx .
Producer’s surplus The excess revenue that a producer has actually received over and above the
lower revenue that it was prepared to accept for the supply of its goods.
Exercise 6.2
1. Evaluate each of the following integrals:
#
3
(a)
#
3
4x2dx (b)
1
2
2
dx
x3
(c)
#
4
1
6
x
# 4x 2 3x 1 4x 1 2 dx
2
(d)
dx
3
2
1
2. Find the exact areas under each of the following curves:
(a) y = 2x2 + x + 3 between x = 1 and x = 5
(b) y = (x − 2)2 between x = 2 and x = 3
(c) y 5 3 x between x = 4 and x = 25
(d) y = ex between x = 0 and x = 1
1
(e) y 5 between x = 1 and x = e
x
3. Evaluate each of the following definite integrals:
(a)
#
2
x3dx
0
(b)
#
2
x3dx
22
By sketching a rough graph of the cube function between and x = −2 and 2, suggest
a reason for your answer to part (b). What is the actual area between the x axis and
the graph of y = x3, over this range?
4. Find the consumer’s surplus at P = 5 for the following demand functions:
(a) P = 25 − 2Q
(b) P 5
10
Q
5. Find the producer’s surplus at Q = 9 for the following supply functions:
(a) P = 12 + 2Q (b) P 5 20 Q 1 15
6. Find the consumer’s surplus for the demand function
P = 50 − 2Q − 0.01Q2
when
(a) Q = 10 (b) Q = 11
➜
510 Chapter 6 Integration
7. Given the demand function
P = −QD2 − 4QD + 68
and the supply function
P = QS2 + 2QS + 12
Find
(a) the consumer’s surplus
(b) the producer’s surplus
assuming pure competition.
8. If the investment flow is
I(t) = 5000t1/4
calculate the capital formation from the end of the second year to halfway through the
fifth year. Give your answer to the nearest whole number.
9. Given the investment flow
I(t) 5 2400 t
(a) Calculate the total capital formation during the first four years.
(b) Find an expression for the annual capital formation during the Nth year and hence
find the first year in which the annual capital formation exceeds $4000.
10. Calculate the present value of a revenue stream for eight years at a constant rate of
$12 000 per year if the discount rate is 7.5%.
11. A company begins its extraction of oil from a newly discovered oil field at t = 0. The
rate of extraction, measured in thousands of barrels per year, is given by
dN
5 30t 2 2 4t 3
dt
Calculate the number of barrels extracted during the first four years of operation.
Exercise 6.2*
1. Evaluate
(a)
#
10
2
5x2 2 4x 1 6dx (b)
21
#
2
1
(2x 1 5) (2x 1 5)
dx
2. (a) Find the consumer’s surplus at Q = 8 for the demand function
P = 50 − 4Q
(b) The producer’s surplus at Q = a for the supply function
P = 6 + 8Q
is known to be 400. Find the value of a.
Section 6.2 Definite integration 511
3. The demand function is given by
P = 74 − QD2
and the supply function is
P = (QS + 2)2
Calculate the consumer’s and producer’s surplus under pure competition.
4. If the supply and demand functions are given by P = Q + 50 and P 5
4000 , respectively,
Q 1 20
find the equilibrium price and quantity, and calculate the consumer’s and producer’s
surplus.
5. If the supply and demand functions are given by P = 20e0.4Q and P = 100e−0.2Q, respec-
tively, find the equilibrium price and quantity, and calculate the consumer’s and producer’s surplus.
6. If the net investment function is given by
I(t) = 100e0.1t
calculate
(a) the capital formation from the end of the second year to the end of the fifth year;
(b) the number of years required before the capital stock exceeds $100 000.
7. Find the expression for capital formation between t = 0 and t = T for the following net
investment functions:
(a) I(t) = Ata
(b) I(t) = Aeat
where A and a are positive constants.
8. Calculate the present value of a continuous revenue stream of $1000 per year if the
discount rate is 5% and the money is paid
(a) for 3 years (b) for 10 years (c) for 100 years (d) in perpetuity
9. The present value of a continuous revenue stream of $5000 per year with a discount
rate of 10% over n years is $25 000. Find the value of n correct to one decimal place.
10. Write down an expression for g′(x) when
x
g(x) 5
# (5t 2 2t)dt
2
2
11. Write down an expression for the present value of a continuous revenue stream for n
years at a constant rate of $S per year if the discount rate is r%.
12. If a firm’s profit function is π = f(Q) and the level of output, Q, varies between a and
b units, then the firm’s average profit is given by
1
b2a
# f (Q)dQ
b
a
Calculate a firm’s average profit when TR = 100(1 − e−0.1Q), TC = 0.1Q2 + 2Q + 1 and
output varies from 3 to 8 units.
13. A newly elected government decides to reduce pollution in major cities by investing in
a green spaces project. The subsequent investment flow is given by
I(t) = 0.6t2 + 10
where t is measured in years from the start of the government term and I is measured
in millions of dollars per year. Five years later, as a result of a change of administration, planned investment is dramatically reduced so that for the following five years,
investment flow J(t) is given by
J(t) = 25e0.4(5−t) for 5 ≤ t ≤ 10
(a) Verify that I(5) = J(5) and sketch a graph of the investment flow for the complete
10-year period.
(b) After how many years does the total capital spent on this project first exceed $110
million? Give your answer to one decimal place.
14. The demand function of a good is given by P 5 80 2 6 Q
(a) Find the change in consumer’s surplus due to a decrease in price from P = 62 to
P = 56.
(b) Sketch a graph of the demand function and shade the area which corresponds to
the change in consumer’s surplus calculated in part (a).
Formal mathematics
In Section 6.2 integration was used to find the area between the x axis, the curve y = f(x) and
the vertical lines x = a and x = b. In all cases the region was bounded on all sides leading to a
finite area. It is sometimes possible to find the area of unbounded regions such as that shown
in Figure 6.7. Integrals representing such areas are referred to as improper integrals. It may
seem surprising that a region which ‘goes on forever’ can actually have a finite area at all.
However, provided the graph approaches the x axis fast enough, the area itself
is finite and has a value which we can find using integration. Formally we define:
#
`
# f (x)dx
N
f (x)dx 5 lim
N
a
`
(assuming the limit exists)
a
We first work out an expression, in terms of N, for the area under the curve between x = a
and x = N, and then see whether or not it converges as N tends to ∞.
y5
1
x3
1
Figure 6.7
Example
Evaluate the following integrals (where possible):
#
`
(a)
1
1
dx
x3
#
`
(b)
1
1
x
dx
Solution
(a) We begin by finding the finite area under the graph of y = x−3 from x = 1 to x = N:
#
N
1
1
x 23 dx 5 2 x 22
2
N
1
1
1
1
1
5 2 N 22 1 5 2
2
2
2
2N 2
514 Chapter 6 Integration
From the definition of an infinite integral we now consider what happens to this expression as N increases without bound.
#
`
1
1
1
1
dx 5 lim
5
2
2
3
N
`
2
2N
2
x
1
because
1
2N 2
0 as N
`
In this case the unbounded region, shown in Figure 6.7, has a finite area of 1/2.
(b) We begin by finding the finite area under the graph of y 5 x21/2 from x = 1 to x = N:
#x
N
21/2
N
dx 5 2x1/2
5 2N 1/2 2 2 5 2 N 2 2
1
1
In this case the term N is not finite as N S ∞, so the integral does not exist.
A similar situation arises when there is ‘leakage’ along the y axis as shown in Figure 6.8.
y
y 5 f(x)
a
x
Figure 6.8
We define
#
# f (x)dx
a
a
f (x)dx 5 lim
N
0
0
N
which may or may not exist, depending on how quickly the curve approaches the y axis.
Example
Evaluate the following integrals (where possible):
(a)
#
1
1
dx (b)
x3
0
#
1
0
1
dx
x
Solution
#
1
1 22
(a)x23 dx 5 2 x
2
N
1
1
1
52 1
which is not finite as N S 0, so the integral does
2
2N 2
N
not exist.
(b)
#
1
N
x
2
1
2
1 1
5 2 2 2 N which converges to 2 as N S 0, so
dx 5 2x 2
N
#
1
0
1
dx 5 2.
x
Multiple choice questions
Question 1
#
1
3
Find the indefinite integral 4x3 2 5x 1 5 2 x 1 2 dx
x
3
A. x4 2 2.5x2 1 5x 2 ln x 1 x 1 c
3
B. x4 2 2.5x2 1 5x 2 ln x 2 x 1 c
2
3
C. x4 2 2.5x2 1 5x 1 x 2 2 1 c
x
1
3
D. x4 2 2.5x2 1 5x 2 2 2 x 1 c
x
3
E. x4 2 2.5x2 1 5x 2 lnx 2 2 1 c
x
Question 2
# 3x 2 4x 1 8dx
3
Evaluate the definite integral
2
2
A. 31
B. 15
C. 12
D. 27
E. 17
Question 3
If the marginal revenue function is given by MR = 100 − 6Q2, find the corresponding
demand function.
A. P = 100Q2 − 2Q3 B. P = 100 − 2Q2 C. P = 100 − 2Q3
D. P = 100Q − 2Q2 E. P = 100Q − 2Q3
Question 4
The supply function of a good is P = 40 + 0.002Q2. Which of the following could be used
to work out the producer’s surplus at (Q0, P0)?
#
Q0
A.
40 1 0.002Q2 dQ 1 P0Q0
0
#
40 1 0.002Q2 dQ 2 P0Q0
0
D. Q0P0 2
#
Q0
40 1 0.002Q2 dQ
0
# 40 1 0.002Q dQ
P0
E.
2
0
Q0
C.
# 40 1 0.002Q dQ
P0
B. Q0P0 1
2
0
516 Chapter 6 Integration
Question 5
If the investment flow of a company is I(t) = 8000t1/3, calculate the capital formation during
the first eight years.
A. 96 000
B. 128 000
C. 76 800
D. 256 000
E. 84 000
Question 6
Calculate the present value of a continuous revenue stream for five years at a constant rate
of $7500 per year if the discount rate is 3.4%.
A. $34 485.70
B. $36 678.54
C. $36 798.45
D. $38 775.00
E. $38 993.63
Question 7
2
where Y is national
Y
income. If consumption is 20 when Y = 0, find the consumption when Y = 16.
The marginal propensity to consume is given by MPC = 0.5 1
A. 28
B. 24
C. 36
D. 44
E. 34
Question 8
The diagram shows the graphs of y = 4 and y = 0.5x2 + 2. Find the shaded area.
y
5
4
3
2
1
1
2
3
x
Multiple choice questions 517
A. 5/3
B. 8/3
C. 4/3
D. 16/3
E. 20/3
Question 9
# 8x 2 6 dx 5 18 where a > 0, find a.
2a
If
a
A. 1.5
B. 2.0
C. 2.5
D. 3
E. 3.5
Question 10
What is the consumer’s surplus at P = 75 when the demand function is P = 100 − Q2?
A. 75/3
B. 125/3
C. 225/3
D. 250/3
E. 100/3
Question 11
Find the value of
A. 3/2
B. 1
C. 4
D. 1/2
E. 2
#
6
2
1
dx
2x 2 3
Examination questions
Question 1
A monopolist has a marginal cost function, 6Q2 + 5Q, and the fixed costs are 25. The
demand function is 2P + Q = 510.
(a) Find an expression for the profit function in terms of Q.
(b) Hence find the production level that maximises profit and explain why this is a
maximum.
Question 2
(a) Find the area under the curve y = x2 + 2x + 3 between x = 1 and x = 2.
(b) Integrate each of the following functions with respect to x:
2
(i) x5 − 2x2 + 9 (ii) ex 1 3 (iii) x 1 3 2 9 x
x
x
(c) Show that
1
2
3x 1 1
1
5
x 2 1 x 1 3 (x 2 1)(x 1 3)
# (x 23x1)(x1 11 3) dx 5 2ln 165
5
Deduce that
2
Question 3
An investment flow is I(t) 5 900 t where t is measured in years.
(a) Calculate the total capital formation during the first four years.
(b) Calculate the total capital formation from the end of the second year to the beginning
of the sixth year. Round your answer to two decimal places.
(c) Find an expression for the total capital formation during the first N years and hence
find the year in which the total capital formation reaches 16 200.
Question 4
A firm’s marginal revenue and marginal cost functions are given by
MR = 140 − 6Q and MC = Q2 + Q + 20
Fixed costs are 10.
(a) Write down an expression for total revenue and deduce the corresponding demand
function.
(b) Write down an expression for the total cost function.
(c) Find the value of Q which maximises profit, π, and state the corresponding value of π.
Examination questions 519
Question 5
The supply and demand functions of a good are given by
P 5 32 1 QS2
P 5 140 2
QD2
3
where P, QS and QD denote the price, quantity supplied and quantity demanded,
­respectively.
(a) Draw sketch graphs of these functions on the same diagram with quantity on the
horizontal axis and price on the vertical axis.
(b) Calculate the producer’s and consumer’s surpluses at the equilibrium point.
(c) If the government imposes a fixed tax on this good explain the effect, if any, on
(i) equilibrium price and quantity;
(ii) consumer’s surplus.
Question 6
A fund is established to provide a continuous revenue stream at a constant rate for a prescribed number of years when the discount rate is 4%.
(a) Calculate the present value if the fund provides $4000 per year for eight years. Give
your answer to the nearest cent.
(b) Determine the number of years that will fund an annual pay-out of $4000 if the present
value is $45 000. Give your answer to the nearest year.
Question 7
(a) Find the following indefinite integrals:
#1 12 6x dx (ii) #x 1 1 3x dx
(i)
2
(b) The supply of a good can be modelled by a linear function, aP + bQ = 1 where a and
b are constants. It is known that when the price, P is 12, the quantity supplied, Q is 6
and the producer’s surplus is 18. Find the values of a and b.
Question 8
(a) Find the total cost function given that the marginal cost function MC = 2Q 1 3 Q and
that total costs are 40 when Q = 4.
(b) The inverse demand function for a good is Q 5
10 2 P
. Find the consumer’s surplus
2P
when the price P = 2. Round your answer to two decimal places.
520 Chapter 6 Integration
Question 9
The price elasticity of demand is given by E 5 2P2 , where P and Q denote price and quan2Q
tity demanded, respectively. It is known that quantity demanded is 5 when the price is 10.
Find an expression for the demand function.
Question 10
(a) Sketch a graph of the supply curve P = 4 + Q2 on the interval 0 ≤ Q ≤ 3.
(b) On the same diagram, sketch the line, P = 13.
(c)Work out the exact area bounded between the curve P = 4 + Q2, the line P = 13 and
the P axis. Explain what this area represents.
Question 11
A hospital trust decides to install a new rooftop solar power system which will save money
on its energy bills. It estimates that t years after installation the savings flow (in $1000s per
year) will steadily increase according to
S(t) = 0.3t + 45.6
The flow of money needed to repair and maintain the system in subsequent years is
modelled by
C(t) = 0.5t2 + 2t
(a) Draw a sketch graph of the savings and cost functions on the same diagram.
(b) Solve the equation S(t) = C(t) to determine the period during which the savings exceed
the maintenance costs, and shade the area which determines the net savings during this
time.
(c) Use integration to find the exact value of net savings in part (b).
Question 12
The demand function of a good is given by P 5
80
Q
3
(a) Show that the price elasticity of demand is a constant.
(b) Sketch a graph of the demand curve.
(c) Shade the area which gives the consumer’s surplus at Q = 64 and use integration to
calculate its value.
This page intentionally left blank
Chapter 7
Matrices
The impression that you may have gained from reading this text is that mathematics consists of
one main topic, calculus, and that every other topic is just a variation on this theme. This is far
from the truth. In this chapter and the next, we look at two refreshingly different branches of
mathematics. It would be useful for you to have studied Chapter 1, although even this is not essen­
tial. There are three sections which need to be read in order.
Section 7.1 introduces the concept of a matrix, which is a convenient mathematical way of repre­
senting information displayed in a table. By defining the matrix operations of addition, subtraction
and multiplication, it is possible to develop an algebra of matrices. Simple economic examples
are used to illustrate these definitions, and it is shown that the rules of matrix manipulation are
almost identical to those of ordinary arithmetic. In Section 7.2 you are shown how to calculate
the inverse of a matrix. This is analogous to the reciprocal of a number and enables matrix equa­
tions to be solved. In particular, inverses provide an alternative way of solving systems of simul­
taneous linear equations and so can be used to solve problems in statics. Section 7.3 describes
Cramer’s rule for solving systems of linear equations. This method is a particularly useful way of
solving economic models where only a selection of endogenous variables need to be determined.
Section 7.1
Basic matrix operations
Objectives
At the end of this section you should be able to:
●
Understand the notation and terminology of matrix algebra.
●
Find the transpose of a matrix.
●
Add and subtract matrices.
●
Multiply a matrix by a scalar.
●
Multiply matrices together.
●
Represent a system of linear equations in matrix notation.
Suppose that a firm produces three types of good, G1, G2 and G3, which it sells to two cus­
tomers, C1 and C2. The monthly sales for these goods are given in Table 7.1.
During the month the firm sells 3 items of G2 to customer C1, 6 items of G3 to customer
C2, and so on. It may well be obvious from the context exactly what these numbers represent.
Under these circumstances it makes sense to ignore the table headings and to write this infor­
mation more concisely as
A5
7 3 4
1 5 6
which is an example of a matrix. Quite generally, any rectangular array of numbers surrounded
by a pair of brackets is called a matrix (plural matrices), and the individual numbers constituting
the array are called entries or elements. In this text we use square brackets, although it is
equally correct to use parentheses (that is, round brackets) instead. It helps to think of a matrix
as being made up of rows and columns. The matrix A has two rows and three columns and
is said to have order 2 × 3. In general, a matrix of order m × n has m rows and n columns.
We denote matrices by capital letters in bold type (that is, A, B, C, . . . ) and their elements
by the corresponding lower-case letter in ordinary type. In fact, we use a rather clever double
subscript notation so that aij stands for the element of A which occurs in row i and column j.
Referring to the matrix A above, we see that
a12 = 3
(row 1 and column 2 of A)
and a23 = 6 (row 2 and column 3 of A)
Table 7.1
Monthly sales for goods
Sold to
customer
C1
C2
G1
G2
G3
7
1
3
5
4
6
Section 7.1 Basic matrix operations 525
A general matrix D of order 3 × 2 would be written
d11 d12
d 21 d 22
d31 d32
Similarly, a 3 × 3 matrix labelled E would be written
e11 e12
e21 e22
e31 e32
e13
e23
e33
Practice Problem
1. Let
A5
1 2
3 4
B 5 1 21 0 6 2
1 0 2 3 1
C5 5 7 9 0 2
3 4 6 7 8
D5 6
(a) State the orders of the matrices A, B, C and D.
(b) Write down (if possible) the values of
a11, a22, b14, c25, c33, c43, d11
All we have done so far is to explain what matrices are and to provide some notation for
handling them. A matrix certainly gives us a convenient shorthand to describe information
presented in a table. However, we would like to go further than this and to use matrices to
solve problems in economics. To do this we describe several mathematical operations that
can be performed on matrices, namely:
●
●
●
●
transposition;
addition and subtraction;
scalar multiplication;
matrix multiplication.
One obvious omission from the list is matrix division. Strictly speaking, it is impossible to
divide one matrix by another, although we can get fairly close to the idea of division by defin­
ing something called an inverse, which we consider in Section 7.2.
Advice
If you have not met matrices before, you might like to split this section into two separate parts. You
are advised to work through the material as far as 7.1.4 now, leaving matrix multiplication for another
session.
526 Chapter 7 Matrices
7.1.1 Transposition
In Table 7.1 the rows correspond to the two customers and the columns correspond to the
three goods. The matrix representation of the table is then
A5
7 3 4
1 5 6
The same information about monthly sales could easily have been presented the other way
round, as shown in Table 7.2. The matrix representation would then be
7 1
B5 3 5
4 6
We describe this situation by saying that A and B are transposes of each other and write
AT = B
read ‘A transpose equals B’
or equivalently
BT = A
read ‘B transpose equals A’
The transpose of a matrix is found by replacing rows by columns, so that the first row becomes
the first column, the second row becomes the second column, and so on. The number of rows
of A is then the same as the number of columns of AT and vice versa. Consequently, if A has
order m × n, then AT has order n × m.
Example
Write down the transpose of the matrices
1 7 0 3
D5 2 4 6 0
5 1 9 2
E5
26
3
Solution
The transpose of the 3 × 4 matrix D is the 4 × 3 matrix
1
7
D 5
0
3
T
2
4
6
0
5
1
9
2
The transpose of the 2 × 1 matrix E is the 1 × 2 matrix
ET = [−6
3]
Table 7.2
Sold to customer
Monthly
sales for
goods
G1
G2
G3
C1
C2
7
3
4
1
5
6
Section 7.1 Basic matrix operations 527
Practice Problem
2. Write down the transpose of the following matrices:
1 4
3 7
A5
2 1
2 25
0
6
3
1
1 2
1 4
5 21
8 0
B5 1 5 7 9
1 2 3
C5 2 4 5
3 5 6
There are two particular shapes of matrices which are given special names. A matrix that
has only one row, such as
c = [5
2
−4]
1
is called a row vector, and a matrix that has only one column, such as
23
10
6
d 5 27
1
9
2
is called a column vector. It is standard practice to identify vectors using lower-case rather
than upper-case letters. In books they are set in bold type. If you are writing them down by
hand, then you should underline the letters and put
c (or possibly ~
c)
and
d (or possibly ~
d)
This is a useful convention since it helps to distinguish scalar quantities such as x, y, a, b,
which denote single numbers, from vector quantities such as x, y, a, b, which denote matrices
with one row or column. Incidentally, it is actually quite expensive to print column vectors
in books and journals since it is wasteful on space, particularly if the number of elements is
large. It is then more convenient to use the transpose notation and write the vector horizont­
ally. For example, the 7 × 1 matrix d given previously would be printed as
d = [−3
10
6 −7
1
9 2]T
where the superscript T tells us that it is the column vector that is intended.
7.1.2 Addition and subtraction
Let us suppose that, for the two-customer three-product example, the matrix
A5
7 3 4
1 5 6
528 Chapter 7 Matrices
gives the sales for the month of January. Similarly, the monthly sales for February might be
given by
B5
6 2 1
0 4 4
This means, for example, that customer C1 buys 7 items of G1 in January and 6 items of G1
in February. Customer C1 therefore buys a total of
7 + 6 = 13
items of G1 during the two months. A similar process can be applied to the remaining goods
and customers, so that the matrix giving the sales for the two months is
C5
5
716 312
110 514
411
614
13 5 5
1 9 10
We describe this by saying that C is the sum of the two matrices A and B and writing
C=A+B
In general, to add (or subtract) two matrices of the same size, we simply add (or subtract) their
corresponding elements. It is obvious from this definition that, for any two m × n matrices, A
and B,
A+B=B+A
because it is immaterial which way round two numbers are added. Note that in order to
combine matrices in this way, it is necessary for them to have the same order. For example,
it is impossible to add the matrices
D5
1 27
1 3
1 2
and E 5 1 1
3 5
because D has order 2 × 2 and E has order 3 × 2.
Example
Let
9 23
A5 4
1
2
0
5 2
and B 5 21 6
3 4
Find
(a) A + B (b) A − B (c) A − A
Section 7.1 Basic matrix operations 529
Solution
5
9 23
(a) A 1 B 5 4
1 1 21
3
2 0
2
14 21
6 5 3
7
4
5
4
5
9 23
(b) A 2 B 5 4
1 2 21
3
2 0
2
4 25
5 25
6 5
21 24
4
0 0
23
1 5 0 0
0 0
0
9 23
9
(c) A 2 A 5 4
1 2 4
2 0
2
The result of part (c) of this example is a 3 × 2 matrix in which every entry is zero. Such
a matrix is called a zero matrix and is written 0. In fact, there are lots of zero matrices, each
corresponding to a particular order. For example,
0 0
0 0
[0]
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
are the 1 × 1, 2 × 2, 4 × 1 and 4 × 6 zero matrices, respectively. However, despite this, we
shall use the single symbol 0 for all of these since it is usually clear in any actual example
what the order is and hence which particular zero matrix is being used. It follows from the
definition of addition and subtraction that, for any matrix A,
A−A=0
A+0=A
The role played by the matrix 0 in matrix algebra is therefore similar to that of the number
0 in ordinary arithmetic.
Practice Problem
3. Let
A5
7 5
2 1
B5
5
4
C5
2
2
D5
26 2
1 29
05
0
0
Find (where possible)
(a) A + D (b) A + C (c) B – C (d) C – 0 (e) D – D
530 Chapter 7 Matrices
7.1.3 Scalar multiplication
Returning to the two-customer three-product example, let us suppose that the sales are the
same each month and are given by
A5
7 3 4
1 5 6
This means, for example, that customer C1 buys 7 items of G1 every month, so in a whole
year C1 buys
12 × 7 = 84
items of G1. A similar process applies to the remaining goods and customers, and the matrix
giving the annual sales is
B5
84 36 48
12 3 7 12 3 3 12 3 4
5
12 60 72
12 3 1 12 3 5 12 3 6
Matrix B is found by scaling each element in A by 12, and we write
B = 12A
In general, to multiply a matrix A by a scalar k, we simply multiply each element of A by k.
If
1 2 3
A5 4 5 6
7 8 9
then
2A 5
2 4 6
8 10 12
14 16 18
21 22 23
2A 5 (21)A 5 24 25 26
27 28 29
0 0 0
0A 5 0 0 0 5 0
0 0 0
In ordinary arithmetic we know that
a(b + c) = ab + ac
for any three numbers a, b and c. It follows from our definitions of matrix addition and scalar
multiplication that
k(A + B) = kA + kB
for any m × n matrices A and B, and scalar k.
Another property of matrices is
k(lA) = (kl)A
for scalars k and l. Again, this follows from the comparable property
Section 7.1 Basic matrix operations 531
a(bc) = (ab)c
for ordinary numbers.
You are invited to check these two matrix properties for yourself in the following
problem.
Practice Problem
4. Let
1 22
A5 3
5
0
4
0 21
7
1
6
and B 5 2
(1) Find
(a) 2A (b) 2B (c) A + B (d) 2(A + B)
Hence verify that
2(A + B) = 2A + 2B
(2) Find
(a) 3A (b) –6A
Hence verify that
−2(3A) = −6A
7.1.4 Matrix multiplication
Advice
Hopefully, you have found the matrix operations considered so far in this section easy to understand.
We now turn our attention to matrix multiplication. If you have never multiplied matrices before, you
may find that it requires a bit more effort to grasp, and you should allow yourself extra time to work
through the problems. There is no need to worry. Once you have performed a dozen or so matrix
multiplications, you will find that the technique becomes second nature, although the process may
appear rather strange and complicated at first sight.
We begin by showing you how to multiply a row vector by a column vector. To illustrate this
let us suppose that goods G1, G2 and G3 sell at $50, $30 and $20, respectively, and let us
introduce the row vector
p = [50
30
20]
If the firm sells a total of 100, 200 and 175 goods of type G1, G2 and G3, respectively, then
we can write this information as the column vector
100
q 5 200
175
532 Chapter 7 Matrices
The total revenue received from the sale of G1 is found by multiplying the price, $50, by the
quantity, 100, to get
$50 × 100 = $5000
Similarly, the revenue from G2 and G3 is
$30 × 200 = $6000
and
$20 × 175 = $3500
respectively. The total revenue of the firm is therefore
TR = $5000 + $6000 + $3500 = $14 500
The value of TR is a single number and can be regarded as a 1 × 1 matrix: that is,
[14 500]
This 1 × 1 matrix is obtained by multiplying together the price vector, p, and the quantity
vector, q, to get
100
200 5 [14 500]
175
[50 30 20]
The value 14 500 is found by multiplying the corresponding elements of p and q and then
adding together: that is,
[50
30
20] 100 5 [5000 1 6000 1 3500] 5 [14 500]
200
175
In general, if a is the row vector
[a11
a12 a13 . . .
a1s]
and b is the column vector
b11
b21
b31
bs1
then we define the matrix product
ab 5 [a11 a12
a13 . . . a1s ]
b11
b21
b31
bs1
Section 7.1 Basic matrix operations 533
to be the 1 × 1 matrix
[a11b11 + a12b21 + a13b31 + . . . + a1sbs1]
It is important to notice that the single element in the 1 × 1 matrix ab is found by multiplying
each element of a by the corresponding element of b. Consequently, it is essential that both
vectors have the same number of elements. In other words, if a has order 1 × s and b has
order t × 1, then it is only possible to form the product ab when s = t.
Example
If
2
5
a 5 [1 2 3 4], b 5
21
0
6
and c 5 9
2
find ab and ac.
Solution
Using the definition of the multiplication of a row vector by a column vector we have
ab 5 [1 2 3 4]
2
5
5 [1(2) 1 2(5) 1 3(21) 1 4(0)] 5 [9]
21
0
We have set out the calculations in this way so that you can see how the value, 9, is
obtained. There is no need for you to indicate this in your own answers and you may
simply write
[1 2 3 4]
2
5
5 [9]
21
0
without bothering to insert any intermediate steps.
It is impossible to multiply a and c because a has four elements and c has only three
elements. You can see the problem if you actually try to perform the calculations, since
there is no entry in c with which to multiply the 4 in a.
[1 2 3 4]
6
9 5 [1(6) 1 2(9) 1 3(2) 1 4(?)]
2
534 Chapter 7 Matrices
Practice Problem
5. Let
0
21
a 5 [1 21 0 3 2], b 5 [1 2 9], c 5 1
1
2
and d 5
22
1
0
Find (where possible)
(a) ac
(b) bd
(c) ad
We now turn our attention to general matrix multiplication, which is defined as follows.
If A is an m × s matrix and B is an s × n matrix, then
C = AB
is an m × n matrix and cij is found by multiplying the ith row of A into the jth column of B.
There are three things to notice about this definition. First, the number of columns of A is
the same as the number of rows of B. Unless this condition is satisfied, it is impossible to form
the product AB. Secondly, the matrix C has order m × n, where m is the number of rows of
A and n is the number of columns of B. Finally, the elements of C are found by multiplying
row vectors by column vectors.
To illustrate the technique consider what happens when we find the product of the two
matrices
A5
2 1 0
1 0 4
3 1 2 1
and B 5 1 0 1 2
5 4 1 1
It is a good idea to check before you begin any detailed calculations that it is possible to
multiply these matrices and also to identify the order of the resulting matrix. In this case
A is a 2 × 3 matrix and
B is a 3 × 4 matrix
Matrix A has three columns and B has the same number of rows, so it is possible to find AB.
Moreover, AB must have order 2 × 4 because A has two rows and B has four columns. Hence
3 1 2 1
c11 c12
1 0 1 2 5
c21 c22
5 4 1 1
2 1 0
1 0 4
c13
c23
c14
c24
All that remains for us to do is to calculate the eight numbers cij.
The number c11 in the top left-hand corner lies in the first row and first column, so to find
its value we multiply the first row of A into the first column of B to get
2
1
1
0
0
4
3
1
5
1
0
4
2 1
1 2
1 1
because 2(3) + 1(1) + 0(5) = 7.
7 c12
c21 c22
c13
c23
c14
c24
Section 7.1 Basic matrix operations 535
The number c12 lies in the first row and second column, so to find its value we multiply the
first row of A into the second column of B to get
2
1
1
0
0
4
3
1
5
1
0
4
2 1
1 2
1 1
7
2
c21 c22
c13
c23
c14
c24
because 2(1) + 1(0) + 0(4) = 2.
The values of c13 and c14 are then found in a similar way by multiplying the first row of A
into the third and fourth columns of B, respectively, to get
2
1
1
0
0
4
3 1
1 0
5 4
2
1
1
1
2
1
7
2
c21 c22
5
c23
4
c24
because 2(2) + 1(1) + 0(1) = 5 and 2(1) + 1(2) + 0(1) = 4.
Finally, we repeat the whole procedure along the second row of C. The elements c21, c22,
c23 and c24 are calculated by multiplying the second row of A into the four columns of B in
succession to get
2
1
1
0
0
4
3
1
5
1
0
4
2
1
1
1
2
1
7
23
2
17
5
6
4
5
because
1(3) + 0(1) + 4(5) = 23
1(1) + 0(0) + 4(4) = 17
1(2) + 0(1) + 4(1) = 6
1(1) + 0(2) + 4(1) = 5
In this example we have indicated how to build up the matrix C in a step-by-step manner
and have used highlights to show you how the calculations are performed. This approach has
been adopted merely as a teaching device. There is no need for you to set your calculations out
in this way, and you are encouraged to write down your answer in a single line of working.
Advice
Take the trouble to check before you begin that it is possible to form the matrix product and to
anticipate the order of the end result. This can be done by jotting down the orders of the original
matrices side by side. The product exists if the inner numbers are the same and the order of the
answer is given by the outer numbers: that is,
A
m3s
B
s3n
equal so can
multiply
answer is m 3 n
➜
536 Chapter 7 Matrices
For example, if A, B and C have orders 3 × 5, 5 × 2 and 3 × 4 respectively, then AB exists and has order
3 × 2 because
A
335
B
532
equal so can
multiply
answer is 3 3 2
but it is impossible to form AC because
A
335
C
334
not equal so
cannot multiply
Practice Problem
6. Write down the order of the matrices
1 2
A5 0 1
3 1
and B 5
1 2
3 4
Hence verify that it is possible to form the matrix product
C = AB
and write down the order of C. Calculate all of the elements of C.
We have already noted that matrix operations have similar properties to those of ordinary
arithmetic. Some particular rules of arithmetic are:
a(b + c) = ab + ac
(distributive law)
(a + b)c = ac + bc
(distributive law)
a(bc) = (ab)c
(associative law)
ab = ba
(commutative law)
An obvious question to ask is whether they have a counterpart in matrix algebra. It turns out
that provided the matrices A, B and C have the correct orders for the appropriate sums and
products to exist, then
A(B + C) = AB + AC
(A + B)C = AC + BC
A(BC) = (AB)C
Section 7.1 Basic matrix operations 537
However, although it is true that
ab = ba
for numbers, this result does not extend to matrices. Even if AB and BA both exist, it is not
necessarily true that
AB = BA
This is illustrated in the following example.
Example
If
A5
1 21
2
1
and B 5
1 3
1 2
evaluate AB and BA.
Solution
It is easy to check that it is possible to form both products AB and BA and that they both
have order 2 × 2. In fact
AB 5
1 21
2 1
BA 5
1 3
1 2
1 3
0 1
5
1 2
3 8
7 2
1 21
5
5 1
2 1
so AB ≠ BA.
There are certain pairs of matrices which do commute (that is, for which AB = BA) and
we shall investigate some of these in the next section. However, these are very much the
exception. We therefore have the ‘non-property’ that, in general,
AB ≠ BA
Practice Problems
7. Let
A5
2 1 1
1
1 1
1 2
, D 5 21 1
5 1 0 , B5 2 , C5
3 1
21 1 4
2 1
1
and E 5
1 2 3
4 5 6
Find (where possible)
(a) AB
(b) BA
(c) CD
(d) DC
(e) AE (f) EA
(g) DE
(h) ED
➜
538 Chapter 7 Matrices
8. Evaluate the matrix product Ax, where
1 4 7
A5 2 6 5
8 9 5
x
and x 5 y
z
Hence show that the system of linear equations
x + 4y + 7z = −3
2x + 6y + 5z = 10
8x + 9y + 5z = 1
can be written as Ax = b where
23
b 5 10
1
We conclude this section by showing you how to express a familiar problem in matrix
notation. Section 1.4 described the method of elimination for solving systems of simultaneous
linear equations. For example, we might want to find values of x and y which satisfy
2x − 5y = 6
7x + 8y = −1
Motivated by the result of Practice Problem 8, we write this as
Ax = b
where
A5
2 25
7
8
x5
x
y
b5
6
21
It is easy to check that this is correct simply by multiplying out Ax to get
2 25
7
8
x
2x 2 5y
5
y
7x 1 8y
and so the matrix equation Ax = b reads
6
2x 2 5y
5
21
7x 1 8y
that is,
2x − 5y = 6
7x + 8y = −1
Quite generally, any system of n linear equations in n unknowns can be written as
Ax = b
Section 7.1 Basic matrix operations 539
where A, x and b are n × n, n × 1 and n × 1 matrices, respectively. The matrix A consists of the
coefficients, the vector x consists of the unknowns and the vector b consists of the right-hand
sides. The definition of matrix multiplication allows us to write a linear system in terms of
matrices, although it is not immediately obvious that there is any advantage in doing so. In
the next section we introduce the concept of a matrix inverse and show you how to use this
to solve systems of equations expressed in matrix form.
Throughout this section we have noted various properties that matrices satisfy. For convenience these are summarised in the next sub-section.
7.1.5 Summary
Provided that the indicated sums and products make sense,
A+B=B+A
A−A=0
A+0=A
k(A + B) = kA + kB
k(lA) = (kl)A
A(B + C) = AB + AC
(A + B)C = AC + BC
A(BC) = (AB)C
We also have the non-property that, in general,
AB ≠ BA
Key Terms
Column vector A matrix with one column.
Elements
The individual numbers inside a matrix. (Also called entries.)
Matrix A rectangular array of numbers, set out in rows and columns, surrounded by a pair of
brackets. (Plural matrices.)
Order (of a matrix)
m × n.
Row vector
The dimensions of a matrix. A matrix with m rows and n columns has order
A matrix with one row.
Transpose (of a matrix) The matrix obtained from a given matrix by interchanging rows and
columns. The transpose of a matrix A is written AT.
Zero matrix A matrix in which every element is zero.
540 Chapter 7 Matrices
Exercise 7.1
1. The monthly sales (in thousands) of burgers (B1) and bites (B2) in three fast-food
restaurants (R1, R2, R3) are as follows:
B1
B2
B1
B2
R1
R2
R3
35
42
27
39
January
13
24
R1
R2
R3
31
25
17
3
29
16
February
(a) Write down two 2 × 3 matrices J and F, representing sales in January and February,
respectively.
(b) By finding J + F, write down the matrix for the total sales over the two months.
(c) By finding J – F, write down the matrix for the difference in sales for the two months.
2. If
2 3 1 9
A5 1 0 5 0
6 7 8 4
1 7 9 6
B5 2 1 0 5
6 4 5 3
work out
(a) 2A (b) 2B (c) 2A + 2B (d) 2(A + B)
Do you notice any connection between your answers to parts (c) and (d)?
3. If A, B and C are matrices with orders, 3 × 3, 2 × 3 and 4 × 2, respectively, which of
the following matrix calculations are possible? If the calculation is possible, state the
order of the resulting matrix
4B,
A + B,
3BT + C,
AB,
BTA,
(CB)T,
CBA
4. A firm manufactures three products, P1, P2 and P3, which it sells to two customers, C1 and
C2. The number of items of each product that are sold to these customers is given by
P1 P2
6
7
A 5 C1
1
C2 2
P3
9
2
The firm charges both customers the same price for each product according to
P1
B = [100
P2
500
P3
200]T
To make each item of type P1, P2 and P3, the firm uses four raw materials, R1, R2, R3
and R4. The number of tonnes required per item is given by
R1
P1 1
C 5 P2 1
P3 0
R2
0
1
0
R3 R4
0
1
2
1
1
1
Section 7.1 Basic matrix operations 541
The cost per tonne of raw materials is
R1 R2 R3 R4
D = [20 10 15 15]T
In addition, let
E = [1
1]
Find the following matrix products and give an interpretation of each one.
(a) AB
(b) AC (c) CD (d) ACD (e) EAB
(f) EACD
(g) EAB – EACD
5. A firm orders 12, 30 and 25 items of goods G1, G2 and G3. The cost of each item of
G1, G2 and G3 is $8, $30 and $15, respectively.
(a) Write down suitable price and quantity vectors, and use matrix multiplication to
work out the total cost of the order.
(b) Write down the new price vector when the cost of G1 rises by 20%, the cost of G2
falls by 10% and the cost of G3 is unaltered. Use matrix multiplication to work out
the new cost of the order and hence find the overall percentage change in total cost.
6. (1) Let
1 2
A5 3 4
5 6
and B 5
1 21
2
1
23 4
Find
(a) AT (b) BT (c) A + B (d) (A + B)T
Do you notice any connection between (A + B)T, AT and BT?
(2) Let
C5
1 4
5 9
and D 5
2 1
21 0
0
1
Find
(a) CT (b) DT (c) CD (d) (CD)T
Do you notice any connection between (CD)T, CT and DT?
7. Verify the equations
(a) A(B + C) = AB + AC (b) (AB)C = A(BC)
in the case when
A5
1 5
5 23
, B5
4 0
2
1
and C 5
21 1
1 2
8. If
1
7
A 5 [1 2 24 3] and B 5
3
2
find AB and BA.
➜
542 Chapter 7 Matrices
9. (a) Evaluate the matrix product Ax, where
A5
7 5
1 3
and x 5
x
y
Hence show that the system of linear equations
7x + 5y = 3
x + 3y = 2
3
.
2
can be written as Ax = b where b 5
(b) The system of equations
2x + 3y − 2z = 6
x − y + 2z = 3
4x + 2y + 5z = 1
can be expressed in the form Ax = b. Write down the matrices A, x and b.
Exercise 7.1*
1. Matrices A, B, C and D have orders 3 × 5, 5 × 2, 5 × 5 and 3 × 5, respectively. State
whether it is possible to perform the following matrix operations.
If it is possible, state the order of the resulting matrix.
(b) (A + C)T (c) A − 2D
(a) 7B
T
T
T
(e) CB (f) D A
(d) BC
T
(g) A + B
2. Two matrices A and B are given by
A5
a21
b
a 1 b 3c 2 b
and B 5
1
3a
2c d 1 1
If A = B, find the values of a, b, c and d.
3. Consider the matrices
A5
a b
d e
c
f
g h
j
k l
and B 5 i
(a) Write down the matrices AT and BT.
(b) Work out the matrix products AB and BTAT.
(c) State the relationship between AB and BTAT, and use this result to simplify
(ATBTCT)T.
Section 7.1 Basic matrix operations 543
4. A chain of sports shops, A, B and C, sells T-shirts, trainers and tennis racquets. The
weekly sales and profit per item are shown in the tables below:
Sales per week
Shop A
Shop B
Shop C
T-shirts
Trainers
Tennis racquets
60
80
10
40
123
0
25
90
25
Profit per item
Shop A ($)
Shop B ($)
Shop C ($)
T-shirts
Trainers
Tennis racquets
1
5
20
1
8
25
1.50
6
30
The 3 × 3 matrices formed from the sales and profit tables are denoted by S and P,
respectively.
(a) If SPT is denoted by A, find the element a11 and give a brief interpretation of this
number.
(b) If STP is denoted by B, find the element b33 and give a brief interpretation of this
number.
5. On a small island there are supermarkets A, L, S and W. In the current year, 30% of
customers buy groceries from A, 20% from L, 40% from S and 10% from W. However,
each year,
A retains 80% of its customers but loses 10% to L, 5% to S and 5% to W.
L retains 90% of its customers but loses 5% to A and 5% to S.
S retains 75% of its customers but loses 10% to A, 10% to L and 5% to W.
W retains 85% of its customers losing 5% to A, 5% to L and 5% to S.
(a) If the original market share is represented by the column vector
0.3
0.2
x5
0.4
0.1
and the matrix representing the transition in supermarket loyalty is
0.8
0.1
T5
0.05
0.05
0.05
0.9
0.05
0
0.1
0.1
0.75
0.05
0.05
0.05
0.05
0.85
work out the matrix product, Tx, and give an interpretation of the elements of the
resulting vector.
(b) Assuming that the same transition matrix applies in subsequent years, work out the
percentage of customers who buy groceries in supermarket L after
(i) two years
6. If A 5
(ii) three years
4 0 7
3 21 4
and B 5
find the matrix X which satisfies the matrix
2 5 1
0 2 1
equation: 2A + XT = 3B.
➜
544 Chapter 7 Matrices
7. Matrices, A, B and C are given by
A5
1 5 0
3 22 4
6 1 0 , B5 4 4 7
2 3 29
25 9 5
3 22 27
5
1
3 0
6
and C 5 24
If D = A(2B + 3C), find d23.
8. Let
A5
I5
1
a b
d 2b
, A21 5
c d
ad 2 bc 2c a
1 0
0 1
and x 5
(ad 2 bc Þ 0)
x
y
Show that
(a) AI = A
and IA = A (b) A−1A = I and AA−1 = I (c) Ix = x
9. For the commodity market:
C = aY + b
and I = cr + d
For the money market:
MS = M*S and
MD = k1Y + k2r + k3
If both markets are in equilibrium, find the matrix, A, such that Ax = b, where x 5
and b 5
Ms* 2 k3
.
b1d
r
Y
Section 7.2
Matrix inversion
Objectives
At the end of this section you should be able to:
●
Write down the 2 × 2 and 3 × 3 identity matrices.
●
Detect whether a matrix is singular or non-singular.
●
Calculate the determinant and inverse of a 2 × 2 matrix.
●
Calculate the cofactors of a 3 × 3 matrix.
●
Use cofactors to find the determinant and inverse of a 3 × 3 matrix.
●
Use matrix inverses to solve systems of linear equations arising in economics.
In this and the following section we consider square matrices, in which the number of rows
and columns are equal. For simplicity we concentrate on 2 × 2 and 3 × 3 matrices, although
the ideas and techniques apply more generally to n × n matrices of any size. We have already
seen that, with one notable exception, the algebra of matrices is virtually the same as the
algebra of numbers. There are, however, two important properties of numbers which we
have yet to consider. The first is the existence of a number, 1, which satisfies
a1 = a
and
1a = a
for any number, a. The second is the fact that corresponding to any non-zero number, a, we
can find another number, a−1, with the property that
a−1a = 1 and
aa−1 = 1
a21 5
1
a
If you have worked through Question 8 of Exercise 7.1* you will know how to extend these to
2 × 2 matrices. In part (a) you showed that, for any 2 × 2 matrix, A,
AI = A and
IA = A
where
I5
1 0
0 1
The matrix I is called the identity matrix and is analogous to the number 1 in ordinary arith­
metic. You also showed in part (b) of Question 8 that corresponding to the 2 × 2 matrix
A5
a b
c d
there is another matrix
A 21 5
1
d 2b
ad 2 bc 2c a
546 Chapter 7 Matrices
with the property that
A−1A = I and AA−1 = I
The matrix A−1 is said to be the inverse of A and is analogous to the reciprocal of a number.
The formula for A−1 looks rather complicated, but the construction of A−1 is in fact very easy.
Starting with some matrix
A5
a b
c d
we first swap the two numbers on the leading diagonal (that is, the elements along the line
joining the top left-hand corner to the bottom right-hand corner of A) to get
d b
c a
swap
a and d
Secondly, we change the sign of the ‘off-diagonal’ elements to get
d 2b
2c a
change signs
of b and c
Finally, we multiply the matrix by the scalar
1
ad 2 bc
to get
1
d 2b
ad 2 bc 2c a
divide each element
by ad 2 bc
The number ad – bc is called the determinant of A and is written as
det(A)
or
A
or
a b
c d
Notice that the last step in the calculation is impossible if
uAu=0
because we cannot divide by zero. We deduce that the inverse of a matrix exists only if the
matrix has a non-zero determinant. This is comparable to the situation in arithmetic where
a reciprocal of a number exists provided the number is non-zero. If the matrix has a non-zero
determinant, it is said to be non-singular; otherwise it is said to be singular.
As an example, consider the matrices
A5
1 2
3 4
and B 5
2 5
4 10
The determinant of the first matrix is
det(A) 5
1 2
5 1(4) 2 2(3) 5 4 2 6 5 22
3 4
Section 7.2 Matrix inversion 547
We see that det(A) ≠ 0, so the matrix is non-singular and the inverse exists. To find A−1 we
swap the diagonal elements, 1 and 4, change the sign of the off-diagonal elements, 2 and 3,
and divide by the determinant, –2. Hence
A21 5 2
1 4 22
22
1
5
3/2 21/2
1
2 23
Of course, if A−1 really is the inverse of A, then A−1A and AA−1 should multiply out to give I.
As a check:
A21A 5
22
3/2
1
21/2
AA21 5
1 2
3 4
22
3/2
1 2
1 0
5
3 4
0 1
1 0
1
5
0 1
21/2
✓
✓
To discover whether or not the second matrix
B5
2 5
4 10
has an inverse, we need to find its determinant.
det(B) 5
2 5
5 2(10) 2 5(4) 5 20 2 20 5 0
4 10
We see that det(B) = 0, so this matrix is singular and the inverse does not exist.
Practice Problem
1. Find (where possible) the inverse of the following matrices. Are these matrices singular
or non-singular?
A5
6 4
1 2
B5
6 4
3 2
One reason for calculating the inverse of a matrix is that it helps us to solve matrix equations
in the same way that the reciprocal of a number is used to solve algebraic equations. We have
already seen in Section 7.1 how to express a system of linear equations in matrix form. Any
2 × 2 system
ax + by = e
cx + dy = f
can be written as
Ax = b
where
A5
a b
c d
x5
x
y
b5
e
f
548 Chapter 7 Matrices
The coefficient matrix, A, and right-hand-side vector, b, are assumed to be given and the
problem is to determine the vector of unknowns, x. Multiplying both sides of
Ax = b
by A−1 gives
A−1(Ax) = A−1b
(A−1A)x = A−1b
(associative property)
Ix = A−1b
(definition of an inverse)
x = A−1b
(Question 8(c) in Exercise 7.1*)
The solution vector x can therefore be found simply by multiplying A−1 by b. We are assum­
ing here that A−1 exists. If the coefficient matrix is singular, then the inverse cannot be found
and the system of linear equations does not possess a unique solution; there are either infin­
itely many solutions or no solution.
Advice
These special cases are dealt with using the elimination method described in Section 1.4. You might
find it instructive to revise both Sections 1.4 and 1.5.
The following example illustrates the use of inverses to find equilibrium prices in supply
and demand models.
Example
The equilibrium prices P1 and P2 for two goods satisfy the equations
−4P1 + P2 = −13
2P1 − 5P2 = −7
Express this system in matrix form and hence find the values of P1 and P2.
Solution
Using the notation of matrices, the simultaneous equations
−4P1 + P2 = −13
2P1 − 5P2 = −7
can be written as
24
1
2 25
P1
213
5
P2
27
that is, as
Ax = b
where
A5
24 1
2 25
x5
P1
P2
b5
213
27
Section 7.2 Matrix inversion 549
The matrix A has determinant
24 1 5 (24)(25) 2 (1)(2) 5 20 2 2 5 18
2 25
To find A−1 we swap the diagonal elements, –4 and –5, change the sign of the off-diagonal
elements, 1 and 2, and divide by the determinant, 18, to get
A21 5
1 25 21
18 22 24
Finally, to calculate x we multiply A−1 by b to get
x 5 A21b 5
1 25 21
18 22 24
1 72
4
213
5
5
54
3
27
18
Hence P1 = 4 and P2 = 3.
Practice Problem
2. The equilibrium prices P1 and P2 for two goods satisfy the equations
9P1 + P2 = 43
2P1 + 7P2 = 57
Express this system in matrix form and hence find the values of P1 and P2.
[You have already solved this particular system in Practice Problem 4 of Section 1.5. You
might like to compare the work involved in solving this system using the method of
elimination described in Chapter 1 and the method based on matrix inverses consid­
ered here.]
Simultaneous equations also arise in macroeconomics.
The equilibrium levels of consumption, C, and income, Y, for the simple two-sector macroeconomic model satisfy the structural equations
Y = C + I*
C = aY + b
where a and b are parameters in the range 0 < a < 1 and b > 0, and I* denotes investment.
The reduced form of the structural equations for this simple model has already been found
in Section 5.3. It is instructive to reconsider this problem using matrices. The objective is to
express the endogenous variables, Y and C, in terms of the exogenous variable I* and para­
meters a and b. The ‘unknowns’ of this problem are therefore Y and C, and we begin by
rearranging the structural equations so that these variables appear on the left-hand sides.
Subtracting C from both sides of
Y = C + I*
gives
Y − C = I*
(1)
550 Chapter 7 Matrices
and if we subtract aY from both sides of
C = aY + b
we get
−aY + C = b
(2)
(It is convenient to put the term involving Y first so that the variables align with those of
equation (1).)
In matrix form, equations (1) and (2) become
1 21
2a 1
Y
I*
5
C
b
that is,
Ax = b
where
1 21
2a 1
A5
x5
Y
C
b5
I*
b
The matrix A has determinant
1 21 5 1(1) 2 (21)(2a) 5 1 2 a
2a 1
which is non-zero because a < 1.
To find A−1, we swap the diagonal elements, 1 and 1, change the sign of the off-diagonal
elements, −1 and −a, and divide by the determinant, 1 − a, to get
A21 5
1
1 1
a
1
12a
Finally, to determine x we multiply A−1 by b to get
x 5 A21b 5
1
1 1
12a a 1
1
I*
I* 1 b
5
b
1 2 a aI* 1 b
Hence
Y5
I* 1 b
aI* 1 b
and C 5
12a
12a
The inverse matrix obviously provides a useful way of solving the structural equations of a
macroeconomic model. In addition, the elements of the inverse matrix can be given an import­
ant economic interpretation. To see this, let us suppose that the investment I* changes by an
amount Δ I* to become I * + Δ I *, with the parameter b held fixed. The new values of Y and
C are obtained by replacing I* by I* + Δ I* in the expressions for Y and C, and are given by
I* 1 DI* 1 b
and
12a
a(I* 1 DI*) 1 b
12a
respectively. The change in the value of Y is therefore
DY 5
I* 1 DI* 1 b I* 1 b
1
DI*
2
5
12a
12a
12a
Section 7.2 Matrix inversion 551
and the change in the value of C is
DC 5
a(I* 1 DI*) 1 b aI* 1 b
a
DI*
2
5
12a
12a
12a
In other words, the changes to Y and C are found by multiplying the change in I* by
1
a
and
12a
12a
respectively. For this reason we call
1
12a
the investment multiplier for Y and
a
12a
the investment multiplier for C.
Now the inverse matrix is
A21 5
1
12a
1
12a
a
12a
1
12a
and we see that these multipliers are precisely the elements that appear in the first column. It
is easy to show, using a similar argument, that the second column contains the multipliers for
Y and C due to changes in the autonomous consumption, b. The four elements in the inverse
matrix can thus be interpreted as follows:
b
I*
Y investment multiplier Y
C investment multiplier C
autonomous consumption multiplier for Y
autonomous consumption multiplier for C
Practice Problem
3. The general linear supply and demand equations for a one-commodity market model
are given by
P = aQS + b
(a > 0, b > 0)
P = −cQD + d
(c > 0, d > 0)
Show that in matrix notation the equilibrium price, P, and quantity, Q, satisfy
1 2a
1 c
P
b
5
Q
d
Solve this system to express P and Q in terms of a, b, c and d. Write down the multiplier
for Q due to changes in b and deduce that an increase in b leads to a decrease in Q.
552 Chapter 7 Matrices
The concepts of determinant, inverse and identity matrices apply equally well to 3 × 3 matri­
ces. The identity matrix is easily dealt with. It can be shown that the 3 × 3 identity matrix is
1 0 0
I5 0 1 0
0 0 1
You are invited to check that, for any 3 × 3 matrix A,
AI = A and
IA = A
Before we can discuss the determinant and inverse of a 3 × 3 matrix, we need to introduce
an additional concept known as a cofactor. Corresponding to each element aij of a matrix A,
there is a cofactor, Aij. A 3 × 3 matrix has nine elements, so there are nine cofactors to be
computed. The cofactor, Aij, is defined to be the determinant of the 2 × 2 matrix obtained
by deleting row i and column j of A (known as a minor), prefixed by a ‘+’ or ‘–’ sign accord­
ing to the following pattern:
1 2 1
2 1 2
1 2 1
For example, suppose we wish to calculate A23, which is the cofactor associated with a23 in
the matrix
a11 a12 a13
A 5 a21 a22 a23
a31 a32 a33
The element a23 lies in the second row and third column. Consequently, we delete the second
row and third column to produce the 2 × 2 matrix
a11 a12
a31 a32
The cofactor, A23, is the determinant of this 2 × 2 matrix prefixed by a ‘–’ sign because from
the pattern
1 2 1
2 1 2
1 2 1
we see that a23 is in a minus position. In other words,
A3 5 2
a11 a12
a31 a32
5 2(a11a32 2 a12a31)
5 2a11a32 1 a12a31
The ‘±’ pattern can also be worked out using (−1)i+j since when i + j is even, (−1)i+j = +1, and
when i + j is odd, (−1)i+j = −1. For the cofactor A23 worked out previously, (−1)2+3 = (−1)5 = −1,
confirming that the element a23 is in the minus position.
Section 7.2 Matrix inversion 553
Example
Find all the cofactors of the matrix
2 4 1
A5 4 3 7
2 1 3
Solution
Let us start in the top left-hand corner and work row by row. For cofactor A11, the element
a11 = 2 lies in the first row and first column, so we delete this row and column to produce
the 2 × 2 matrix
3 7
1 3
Cofactor A11 is the determinant of this 2 × 2 matrix, prefixed by a ‘+’ sign because from
the pattern
1 2 1
2 1 2
1 2 1
we see that a11 is in a plus position. Hence
A11 5 1 3 7
1 3
5 1(3(3) 2 7(1))
5927
52
For cofactor A12, the element a12 = 4 lies in the first row and second column, so we delete
this row and column to produce the 2 × 2 matrix
4 7
2 3
Cofactor A12 is the determinant of this 2 × 2 matrix, prefixed by a ‘–’ sign because from
the pattern
1 2 1
2 1 2
1 2 1
we see that a12 is in a minus position. Hence
A12 5 2 4 7
2 3
5 2(4(3) 2 7(2))
5 2(12 2 14)
52
➜
554 Chapter 7 Matrices
We can continue in this way to find the remaining cofactors:
A13 5 1 4 3 5 22
2 1
A21 5 2 4 1 5 211
1 3
A22 5 1 2 1 5 4
2 3
A23 5 2 2 4 5 6
2 1
A31 5 1 4 1 5 25
3 7
A32 5 2 2 1 5 210
4 7
A33 5 1 2 4 5 210
4 3
Practice Problem
4. Find all the cofactors of the matrix
1 3 3
A5 1 4 3
1 3 4
We are now in a position to describe how to calculate the determinant and inverse of a
3 × 3 matrix. The determinant is found by multiplying the elements in any one row or column
by their corresponding cofactors and adding together. It does not matter which row or column
is chosen; exactly the same answer is obtained in each case. If we expand along the first row
of the matrix
a11
A 5 a21
a31
a12
a22
a32
a13
a23
a33
we get
det(A) = a11A11 + a12A12 + a13A13
Similarly, if we expand down the second column, we get
det(A) = a12A12 + a22A22 + a32A32
The fact that we get the same answer irrespective of the row and column that we use for expan­
sion is an extremely useful property. It provides us with an obvious check on our calculations.
Also, there are occasions when it is more convenient to expand along certain rows or columns
than others.
Section 7.2 Matrix inversion 555
Example
Find the determinants of the following matrices:
2 4 1
A5 4 3 7
2 1 3
and B 5
10 7 5
0 2 0
2 7 3
Solution
We have already calculated all nine cofactors of the matrix
2 4 1
A5 4 3 7
2 1 3
in the previous example. It is immaterial which row or column we use. Let us choose the
second row. The cofactors corresponding to the three elements 4, 3, 7 in the second row were
found to be –11, 4, 6, respectively. Consequently, if we expand along this row, we get
2 4 1
4 3 7 5 4(211) 1 3(4) 1 7(6) 5 10
2 1 3
As a check, let us also expand down the third column. The elements in this column are 1,
7, 3 with cofactors –2, 6, –10, respectively. Hence, if we multiply each element by its cofac­
tor and add, we get
1(−2) + 7(6) + 3(−10) = 10
which is the same as before. If you are interested, you might like to confirm for yourself
that the value of 10 is also obtained when expanding along rows 1 and 3, and down col­
umns 1 and 2.
The matrix
B5
10 7 5
0 2 0
2 7 3
is entirely new to us, so we have no prior knowledge about its cofactors. In general, we
need to evaluate all three cofactors in any one row or column to find the determinant of a
3 × 3 matrix. In this case, however, we can be much lazier. Observe that all but one of the
elements in the second row are zero, so when we expand along this row, we get
det(B) = b21B21 + b22B22 + b23B23
= 0B21 + 2B22 + 0B23
= 2B22
Hence B22 is the only cofactor that we need to find. This corresponds to the element in the
second row and second column, so we delete this row and column to produce the 2 × 2
matrix
10 5
2 3
➜
556 Chapter 7 Matrices
The element b22 is in a plus position, so
B22 5 1 10 5 5 20
2 3
Hence,
det(B) = 2B22 = 2 × 20 = 40
Practice Problem
5. Find the determinants of
1 3 3
A5 1 4 3
1 3 4
270 2372 0
201 0
999
413 0
and B 5 552
[Hint: you might find your answer to Practice Problem 4 useful when calculating the
determinant of A.]
The inverse of the 3 × 3 matrix
a11 a12
A 5 a21 a22
a31 a32
a13
a23
a33
is given by
A
21
1
5
A
A11
A12
A13
A21
A22
A23
A31
A32
A33
Once the cofactors of A have been found, it is easy to construct A−1. We first stack the cofac­
tors in their natural positions
A11
A21
A31
A12
A22
A32
A13
A23
A33
called the
adjugate
matrix
Secondly, we take the transpose to get
A11
A12
A13
A21
A22
A23
A31
A32
A33
called the
adjoint
matrix
Finally, we multiply by the scalar
1
A
Section 7.2 Matrix inversion 557
to get
A
21
1
5
A
A11
A12
A13
A21
A22
A23
A31
A32
A33
divide each
element by
the determinant
The last step is impossible if
A 50
because we cannot divide by zero. Under these circumstances, the inverse does not exist and
the matrix is singular.
Advice
It is a good idea to check that no mistakes have been made by verifying that
A−1A = I and
AA−1 = I
Example
Find the inverse of
2 4 1
A5 4 3 7
2 1 3
Solution
The cofactors of this particular matrix have already been calculated as
A11 = 2,
A12 = 2,
A13 = −2
A21 = −11,
A22 = 4,
A23 = 6
A31 = 25,
A32 = −10,
A33 = −10
Stacking these numbers in their natural positions gives the adjugate matrix
2
2 22
211
4
6
25 210 210
The adjoint matrix is found by transposing this to get
2 211 225
2
4 210
22
6 210
In the previous example the determinant was found to be 10, so
A21 5
1/5 211/10 5/2
2 211
25
1
2
4 210 5 1/5
2/5 21
10 22
6 210
21/5
3/5 21
➜
558 Chapter 7 Matrices
As a check:
A21A 5
1/5 211/10 5/2
1/5
2/5
21
21
21/5
3/5
2 4 1
AA21 5 4 3 7
2 1 3
2 4 1
1 0 0
4 3 7 5 0 1 0 5I ✓
2 1 3
0 0 1
1 0 0
1/5 211/10 5/2
1/5
2/5
21 5 0 1 0 5 I ✓
21/5
0 0 1
3/5
21
Practice Problem
6. Find (where possible) the inverses of
1 3 3
A5 1 4 3
1 3 4
270 2372 0
201 0
999
413 0
and B 5 552
[Hint: you might find your answers to Practice Problems 4 and 5 useful.]
Inverses of 3 × 3 matrices can be used to solve systems of three linear equations in three
unknowns. The general system
a11x + a12 y + a13z = b1
a21x + a22 y + a23z = b2
a31x + a32 y + a33z = b3
can be written as
Ax = b
where
a11 a12
a
A 5 21 a22
a31 a32
a13
a23
a33
x
x5 y
z
b1
b 5 b2
b3
The vector of unknowns, x, can be found by inverting the coefficient matrix, A, and multiply­
ing by the right-hand-side vector, b, to get
x = A−1b
Example
Determine the equilibrium prices of three interdependent commodities that satisfy
2P1 + 4P2 + P3 = 77
4P1 + 3P2 + 7P3 = 114
2P1 + P2 + 3P3 = 48
Section 7.2 Matrix inversion 559
Solution
In matrix notation this system of equations can be written as
Ax = b
where
2 4 1
A5 4 3 7
2 1 3
P1
x 5 P2
P3
77
b 5 114
48
The inverse of the coefficient matrix has already been found in the previous example and is
A21 5
1/5 211/10 5/2
21
1/5
2/5
21/5
21
3/5
so
P1
1/5 211/10 5/2
P2 5 1/5
21
2/5
P3
21/5
21
3/5
77
10
114 5 13
48
5
The equilibrium prices are therefore given by
P1 = 10,
P2 = 13, P3 = 5
Practice Problem
7. Determine the equilibrium prices of three interdependent commodities that satisfy
P1 + 3P2 + 3P3 = 32
P1 + 4P2 + 3P3 = 37
P1 + 3P2 + 4P3 = 35
[Hint: you might find your answer to Practice Problem 6 useful.]
Throughout this section, we have concentrated on 2 × 2 and 3 × 3 matrices. The method
described can be extended to larger matrices of order n × n. The cofactor Aij would be the
determinant of the (n − 1) × (n − 1) matrix left when row i and column j are deleted prefixed
by (−1)i+j. However, the cofactor approach is very inefficient. The amount of working rises
dramatically as n increases, making this method inappropriate for large matrices. The pre­
ferred method for solving simultaneous equations is based on the elimination idea that we
described in Section 1.4. This is easily programmed, and a computer can solve large systems
of equations in a matter of seconds.
560 Chapter 7 Matrices
Key Terms
Cofactor The cofactor of the element, aij, is the determinant of the matrix left when row i and
column j are deleted, multiplied by +1 or −1, depending on whether i + j is even or odd, respectively.
Determinant A determinant can be expanded as the sum of the products of the elements in any
one row or column and their respective cofactors.
Identity matrix An n × n matrix, I, in which every element on the main diagonal is 1 and the other
elements are all 0. If A is any n × n matrix, then AI = A = IA.
A matrix, A−1, with the property that A−1A = I = AA−1.
Inverse matrix
Minor The name given to the cofactor before the ‘±’ pattern is imposed.
Non-singular matrix A square matrix with a non-zero determinant.
Singular matrix A square matrix with a zero determinant. A singular matrix fails to possess an
inverse.
Square matrix A matrix with the same number of rows as columns.
Exercise 7.2
1. (a) Find the determinant of
(i) 2 7
1 4
(ii)
5 6
3 4
(iii) 22 210
1
4
(iv) 26 24
28 27
(b) Find the inverse of each matrix in part (a).
2. Let
A5
2 1
5 1
and B 5
1 0
2 4
(1) Find
(a) u A u (b) u B u (c) u AB u
Do you notice any connection between u A u, u B u and u AB u?
(2) Find
(a) A−1 (b) B−1 (c) (AB)−1
Do you notice any connection between A−1, B−1 and (AB)−1?
3. If the matrices
2 21
3 a
and
2 b
3 24
are singular, find the values of a and b.
4. Evaluate the matrix product,
5 23
210 8
8 3 .
10 5
Hence, or otherwise, write down the inverse of
8 3
.
10 5
Section 7.2 Matrix inversion 561
5. Use matrices to solve the following pairs of simultaneous equations:
(a) 3x + 4y = −1
(b) x + 3y = 8
4x − y = 6
5x − y = 6
6. The demand and supply functions for two interdependent goods are given by
QD1 = 50 − 2P1 + P2
QD2 = 10 + P1 − 4P2
QS1 = −20 + P1
QS2 = −10 + 5P2
(a) Show that the equilibrium prices satisfy
P1
70
P2 5 20
3 21
21
9
(b) Find the inverse of the 2 × 2 matrix in part (a) and hence find the equilibrium
prices.
7. If a, b and k are non-zero, show that
(a) each of these 2 × 2 matrices is singular:
(i) a 0
b 0
(ii)
a
(iii) 1
b
a b
ka kb
b
1
a
(b) each of these 2 × 2 matrices is non-singular:
(i) a b
0 k
0 a
2a 0
(iii)
and B 5
a 4
2 b
(ii)
a b
2b a
Exercise 7.2*
1. If the matrices
A5
1 2
a b
are both singular, determine all possible values of a and b.
2. (a) If
A5
a b
c d
and B 5
e
g
f
h
work out the matrix product, AB.
(b) Hence, show that det(AB) = det(A) × det(B)
562 Chapter 7 Matrices
(c) If A is singular and B is non-singular, what, if anything, can be deduced about AB?
Give a brief reason for your answer.
3. Which one of the following matrices has an inverse which is not listed?
A5
1 1
1 0
0
1
1 21
21 0
, B5
, C5
, D5
, E5
1 0
0 1
1 21
21
0
0 1
4. Find the determinant of the matrix
5 22
3
4 21 25
6
7
9
5. Find the cofactor, A23, of the matrix
5 22
7
1 29
A5 6
8
4 23
6. Find (where possible) the inverse of the matrices
A5
2 1 21
1 3
2
21 2
1
and B 5
1 4 5
2 1 3
21 3 2
Are these matrices singular or non-singular?
7. For the commodity market
C = aY + b
I = cr + d
(0 a < 1, b > 0)
(c < 0, d > 0)
For the money market
MS = M*S
MD = k1Y + k2r + k3 (k1, k3 > 0, k2 < 0)
Show that when the commodity and money markets are both in equilibrium, the
income, Y, and interest rate, r, satisfy the matrix equation
1 2 a 2c
k1
k2
b1d
Y
5 M* 2 k
r
s
3
and solve this system for Y and r. Write down the multiplier for r due to changes in M*S
and deduce that interest rates fall as the money supply grows.
8. Find the determinant of the matrix
2 1 3
A5 1 0 a
3 1 4
in terms of a. Deduce that this matrix is non-singular provided a ≠ 1 and find A−1 in this
case.
Section 7.2 Matrix inversion 563
9. Find the inverse of
22
2
1
1 25 21
2 21 26
Hence find the equilibrium prices of the three-commodity market model given in
Question 6 of Exercise 1.5*.
10. Find the inverse of the matrix
6 3 a
A5 5 4 2
7 2 3
in terms of a.
For what value of a will simultaneous equations of the form
6x + 3y + az = b
5x + 4y + 2z = c
7x + 2y + 3z = d
fail to possess a unique solution?
11. (a) Multiply out the brackets in the expression (a − b)(a − c)(c − b).
(b) Show that the determinant of the matrix
A5
1
a
a2
1
b
b2
1
c
c2
is (a − b)(a − c)(c − b) and deduce that the simultaneous equations
x+y+z=l
ax + by + cz = m
a x + b2y + c2z = n
2
have a unique solution provided, a, b, and c are distinct.
Section 7.3
Cramer’s rule
Objectives
At the end of this section you should be able to:
●
Appreciate the limitations of using inverses to solve systems of linear equations.
●
Use Cramer’s rule to solve systems of linear equations.
●
Apply Cramer’s rule to analyse static macroeconomic models.
●
Apply Cramer’s rule to solve two-country trading models.
In Section 7.2 we described the mechanics of calculating the determinant and inverse of 2 × 2
and 3 × 3 matrices. These concepts can be extended to larger systems in an obvious way,
although the amount of effort needed rises dramatically as the size of the matrix increases.
For example, consider the work involved in solving the system
1
21
0
2
0
2 3
5
4 1
7 23 6
4
5 1
x1
21
x2
1
x3 5 224
x4
15
using the method of matrix inversion. In this case the coefficient matrix has order 4 × 4 and
so has 16 elements. Corresponding to each of these elements is a cofactor. This is defined to
be the 3 × 3 determinant obtained by deleting the row and column containing the element,
prefixed by a ‘+’ or ‘–’ according to the following pattern:
1
2
1
2
2
1
2
1
1
2
1
2
2
1
2
2
Determinants are found by expanding along any one row or column, and inverses are found
by stacking cofactors as before. However, given that there are 16 cofactors to be calculated,
even the most enthusiastic student is likely to view the prospect with some trepidation. To
make matters worse, it frequently happens in economics that only a few of the variables xi
are actually needed. For instance, it could be that the variable x3 is the only one of interest.
Under these circumstances, it is clearly wasteful expending a large amount of effort calculat­
ing the inverse matrix, particularly since the values of the remaining variables, x1, x2 and x4,
are not required.
In this section we describe an alternative method that finds the value of one variable at a
time. This new method requires less effort if only a selection of the variables is required. It is
Section 7.3 Cramer’s rule 565
known as Cramer’s rule and makes use of matrix determinants. Cramer’s rule for solving any
n × n system, Ax = b, states that the ith variable, xi, can be found from
xi 5
det(A i)
det(A)
where Ai is the n × n matrix found by replacing the ith column of A by the right-hand-side
vector b. To understand this, consider the simple 2 × 2 system
x1
26
x2 5 12
7 2
4 5
and suppose that we need to find the value of the second variable, x2, say. According to
Cramer’s rule, this is given by
x2 5
det(A 2)
det(A)
where
A5
7 2
4 5
and A 2 5
7 26
4 12
Notice that x2 is given by the quotient of two determinants. The one on the bottom is that of
the original coefficient matrix A. The one on the top is that of the matrix found from A by
replacing the second column (since we are trying to find the second variable) by the righthand-side vector
26
12
In this case the determinants are easily worked out to get
det(A 2) 5
7 26
5 7(12) 2 (26)(4) 5 108
4 12
det(A) 5
7 2
5 7(5) 2 2(4) 5 27
4 5
Hence
x2 5
108
54
27
Example
Solve the system of equations
1 2 3
24 1 6
2 7 5
x1
9
x2 5 29
x3
13
using Cramer’s rule to find x1.
➜
566 Chapter 7 Matrices
Solution
Cramer’s rule gives
x1 5
det(A1)
det(A)
where A is the coefficient matrix
1 2 3
24 1 6
2 7 5
and A1 is constructed by replacing the first column of A by the right-hand-side vector
9
29
13
which gives
9 2 3
A1 5 29 1 6
13 7 5
If we expand each of these determinants along the top row, we get
9 2 3
det(A1) 5 29 1 6
13 7 5
5 9 1 6 2 2 29 6 1 3 29 1
7 5
13 5
13 7
5 9(237) 2 2(2123) 1 3(276)
5 2315
and
1 2 3
det(A) 5 24 1 6
2 7 5
5 1 1 6 2 2 24 6 1 3 24 1
2 7
7 5
2 5
5 1(237) 2 2(232) 1 3(230)
5 263
Hence
x1 5
det(A1) 2315
55
5
det(A)
263
Section 7.3 Cramer’s rule 567
Practice Problem
1. (a) Solve the system of equations
2x1 + 4x2 = 16
3x1 − 5x2 = −9
using Cramer’s rule to find x2.
(b) Solve the system of equations
4x1 + x2 + 3x3 = 8
−2x1 + 5x2 + x3 = 4
3x1 + 2x2 + 4x3 = 9
using Cramer’s rule to find x3.
We now illustrate the use of Cramer’s rule to analyse economic models. We begin by con­
sidering the three-sector macroeconomic model involving government activity.
Advice
The incorporation of government expenditure and taxation into the model has already been consid­
ered in Section 5.3, and you might like to compare the working involved in the two approaches.
Example
The equilibrium levels of income, Y, disposable income, Yd, and taxation, T, for a threesector macroeconomic model satisfy the structural equations
Y = C + I* + G*
C = aYd + b
(0 < a < 1, b > 0)
Yd = Y − T
T = tY + T*
(t < 1, T* > 0)
Show that this system can be written as Ax = b, where
1 21
0 0
0 1 2a 0
A5
21 0
1 1
2t 0
0 1
Y
C
x5
Yd
T
I* 1 G*
b
b5
0
T*
Use Cramer’s rule to solve this system for Y.
➜
568 Chapter 7 Matrices
Solution
In this model the endogenous variables are Y, C, Yd and T, so we begin by manipulating
the equations so that these variables appear on the left-hand sides. Moreover, since the
vector of ‘unknowns’, x, is given to be
Y
C
Yd
T
we need to arrange the equations so that the variables appear in the order Y, C, Yd, T. For
example, in the case of the third equation
Yd = Y − T
we first subtract Y and add T to both sides to get
Yd − Y + T = 0
but then reorder the terms to obtain
−Y + Yd + T = 0
Performing a similar process with the remaining equations gives
Y − C = I* + G*
C − aYd = b
−Y + Yd + T = 0
−tY + T = T*
so that in matrix form they become
1 21
0 0
0 1 2a 0
21 0
1 1
2t 0
0 1
Y
I* 1 G*
C
b
5
Yd
0
T
T*
The variable Y is the first, so Cramer’s rule gives
Y5
det(A1)
det(A)
where
0
I* 1 G* 21
1 2a
b
A1 5
0
0
1
0
0
T*
and
1 21
0
0
1 2a
A5
1
21
0
2t
0
0
0
0
1
1
0
0
1
1
Section 7.3 Cramer’s rule 569
The calculations are fairly easy to perform, in spite of the fact that both matrices are 4 × 4,
because they contain a high proportion of zeros. Expanding A1 along the first row gives
1 2a 0
b 2a 0
det(A1) 5 (I* 1 G*) 0 1 1 2(21) 0
1 1
0 0 1
T* 0 1
along the first
row the pattern
is 1 2 1 2
Notice that there is no point in evaluating the last two cofactors in the first row, since the
corresponding elements are both zero.
For the first of these 3 × 3 determinants we choose to expand down the first column,
since this column has only one non-zero element. This gives
1 2a 0
1 1 51
0
1 1 5 (1)
0 1
0
0 1
It is immaterial which row or column we choose for the second 3 × 3 determinant, since
they all contain two non-zero elements. Working along the first row gives
b 2a 0
1 1 2(2a) 0 1 5 b 2 aT*
0
1 1 5b
T* 1
0 1
T* 0 1
Hence
det(A1) = (I* + G*)(1) − (−1)(b − aT*) = I* + G* + b − aT*
A similar process can be applied to matrix A. Expanding along the top row gives
0 2a 0
1 2a 0
det(A) 5 (1) 0
1 1
1 1 2 (21) 21
2t
0 1
0
0 1
The first of these 3 × 3 determinants has already been found to be 1 in our previous calcu­
lations. The second 3 × 3 determinant is new, and if we expand this along the first row,
we get
0 2a 0
21 1
5 a(21 1 t)
21
1 1 5 2(2a)
2t 1
2t
0 1
Hence
det(A) = (1)(1) − (−1)a(−1 + t) = 1 − a + at
Finally, we use Cramer’s rule to deduce that
Y5
I* 1 G* 1 b 2 aT*
1 2 a 1 at
570 Chapter 7 Matrices
Practice Problem
2. Use Cramer’s rule to solve the following system of equations for Yd.
1 21 0
0 1 2a
21 0
1
0
2t 0
0
0
1
1
Y
I* 1 G*
C
b
5
Yd
0
T
T*
[Hint: the determinant of the coefficient matrix has already been evaluated in the previ­
ous worked example.]
We conclude this section by introducing foreign trade into our model. In all of our previ­
ous macroeconomic models we have implicitly assumed that the behaviour of different coun­
tries has no effect on the national income of the other countries. In reality this is clearly not
the case, and we now investigate how the economies of trading nations interact. To simplify
the situation we shall ignore all government activity and suppose that there are just two
countries, labelled 1 and 2, trading with each other but not with any other country. We shall
use an obvious subscript notation so that Y1 denotes the national income of country 1, C2
denotes the consumption of country 2 and so on. In the absence of government activity, the
equation defining equilibrium in country i is
Yi = Ci + Ii + Xi − Mi
where Ii is the investment of country i, Xi is the exports of country i and Mi is the imports of
country i. As usual, we shall assume that Ii is determined exogenously and takes a known
value I*i.
Given that there are only two countries, which trade between themselves, the exports of
one country must be the same as the imports of the other. In symbols we write
X1 = M2 and
X2 = M1
We shall assume that imports are a fraction of national income, so that
Mi = miYi
where the marginal propensity to import, mi, satisfies 0 < mi < 1.
Once expressions for Ci and Mi are given, we can derive a system of two simultaneous
equations for the two unknowns, Y1 and Y2, which can be solved either by using Cramer’s
rule or by using matrix inverses.
Example
The equations defining a model of two trading nations are given by
Y1 = C1 + I *1 + X1 − M1
Y2 = C2 + I*2 + X2 − M2
C1 = 0.8Y1 + 200
C2 = 0.9Y2 + 100
M1 = 0.2Y1
M2 = 0.1Y2
Express this system in matrix form and hence write Y1 in terms of I *1 and I*2.
Write down the multiplier for Y1 due to changes in I*2 and hence describe the effect on
the national income of country 1 due to changes in the investment in country 2.
Section 7.3 Cramer’s rule 571
Solution
In this problem there are six equations for six endogenous variables, Y1, C1, M1 and Y2,
C2, M2. However, rather than working with a 6 × 6 matrix, we perform some preliminary
algebra to reduce it to only two equations in two unknowns. To do this we substitute the
expressions for C1 and M1 into the first equation to get
Y1 = 0.8Y1 + 200 + I*
1 + X1 − 0.2Y1
Also, since X1 = M2 = 0.1Y2, this becomes
Y1 = 0.8Y1 + 200 + I*
1 + 0.1Y2 − 0.2Y1
which rearranges as
0.4Y1 − 0.1Y2 = 200 + I*
1
A similar procedure applied to the second set of equations for country 2 gives
−0.2Y1 + 0.2Y2 = 100 + I*2
In matrix form this pair of equations can be written as
0.4 20.1
20.2
0.2
Y1
200 1 I1*
Y2 5 100 1 I2*
Cramer’s rule gives
Y1 5
200 1 I1*
100 1 I2*
20.1
0.2
0.4 20.1
20.2
0.2
5
50 1 0.2I1* 1 0.1I2*
0.06
To find the multiplier for Y1 due to changes in I*2 we consider what happens to Y1 when
I*2 changes by an amount ΔI*2. The new value of Y1 is obtained by replacing I*2 by I*2 + Δ I *2
to get
50 1 0.2I1* 1 0.1(I2* 1 DI2*)
0.06
so the corresponding change in Y1 is
50 1 0.2I1* 1 0.1(I2* 1 DI2*) 50 1 0.2I1* 1 0.1I2*
2
0.06
0.06
0.1
DI *
5
0.06 2
5
5 DI2*
3
DY1 5
We deduce that any increase in investment in country 2 leads to an increase in the national
income in country 1. Moreover, because 5/3 > 1, the increase in national income is greater
than the increase in investment.
572 Chapter 7 Matrices
Practice Problem
3. The equations defining a model of two trading nations are given by
Y1 = C1 + I*1 + X1 − M1
Y2 = C2 + I*2 + X2 − M2
C1 = 0.7Y1 + 50
C2 = 0.8Y2 + 100
I *1 = 200
I*2 = 300
M1 = 0.3Y1
M2 = 0.1Y2
Express this system in matrix form and hence find the values of Y1 and Y2. Calculate
the balance of payments between these countries.
Key Term
Cramer’s rule A method of solving simultaneous equations, Ax = b, by the use of determinants.
The ith variable xi can be computed using det(Ai)/det(A) where Ai is the determinant of the matrix
obtained from A by replacing the ith column by b.
Exercise 7.3
1. (a) Evaluate each of these determinants.
(i) 4 2
1 3
(ii) 27 2
5 3
(iii) 4 27
1
5
(b) Use your answers to part (a) to write down the solution of the simultaneous equations
4x + 2x = −7
x + 3y = 5
2. Use Cramer’s rule to find the value of x which satisfies each of the following pairs of
simultaneous equations:
(a) 7x − 3y = 4
(b) −3x + 4y = 5
2x + 5y = 7
2x + 5y = 12
(c) x + 4y = 9
2x − 7y = 3
3. Use Cramer’s rule to find the value of y which satisfies each of the following pairs of
simultaneous equations:
(a) x + 3y = 9
2x − 4y = −2
(b) 5x − 2y = 7
2x + 3y = −1
(c) 2x + 3y = 7
3x − 5y = 1
4. Use Cramer’s rule to solve the following sets of simultaneous equations:
(a) 4x + 3y = 1
2x + 5y = −3
(b) 4x + 3y = 1
2x + 5y = 11
(c) 4x + 3y = −2
2x + 5y = −36
Section 7.3 Cramer’s rule 573
5. The demand and supply functions for two interdependent goods are given by
QD1 = 400 − 5P1 − 3P2
QD2 = 300 − 2P1 − 3P2
QS1 = − 60 + 3P1
QS2 = −100 + 2P2
(a) Show that the equilibrium prices satisfy
8 3
2 5
P1
460
P2 5 400
(b) Use Cramer’s rule to find the equilibrium price of good 1.
6. Consider the two-sector macroeconomic model
Y = C + I*
C = aY + b
(a) Express this system in the form
Ax = b
where x 5 Y and A and b are 2 × 2 and 2 × 1 matrices to be stated.
C
(b) Use Cramer’s rule to solve this system for C.
7. A total revenue function may be modelled by TR = aQ + bQ2.
(a) If TR = 14 when Q = 2 and TR = 9 when Q = 3, write down a pair of simultaneous
equations for the parameters a and b.
(b) Use Cramer’s rule to solve the equations in part (a) and hence find the total revenue
when Q = 1.
Exercise 7.3*
1. Use Cramer’s rule to solve
3 22 4
(a) 1 4 0
5 7 0
x
11
y 5 9
19
z
for x.
4 5 0
(b) 21 2 3
6 21 2
for y.
x
0
19
y 5
z
230
➜
574 Chapter 7 Matrices
(c)
x
243
y 5
0
z
34
4 28 2
1
0 6
23
6 2
for z.
1 21
0 0
0
1
2a
0 has determinant, 1 – a + at.
2. The matrix
21 0
1 1
2t 0
0 1
Use Cramer’s rule to solve the following system of equations for C:
1 21
0 0
0 1 2a 0
21 0
1 1
0 1
2t 0
Y
I* 1 G*
C
b
5
Yd
0
T
T*
3. Consider the three-sector macroeconomic model:
Y = C + I* + G*
C = a(Y − T) + b
T = tY + T*
(a) Express this system in the form
Ax = b
Y
where x 5 C and A and b are 3 × 3 and 3 × 1 matrices to be stated.
T
(b) Use Cramer’s rule to solve this system for Y.
4. Consider the macroeconomic model defined by
Y = C + I* + G* + X* − M
C = aY + b
(0 < a < 1, b > 0)
M = mY + M*
(0 < m < 1, M* > 0)
Show that this system can be written as Ax = b, where
1 21
A 5 2a 1
2m 0
1
0
1
Y
x5 C
M
b5
I* 1 G* 1 X*
b
M*
Use Cramer’s rule to show that
Y5
b 1 I* 1 G* 1 X* 2 M*
12a1m
Write down the autonomous investment multiplier for Y and deduce that Y increases
as I* increases.
Section 7.3 Cramer’s rule 575
5. Consider the macroeconomic model defined by
national income:
Y = C + I + G*
(G* > 0)
consumption:
C = aY + b
(0 < a < 1, b > 0)
investment:
I = cr + d
(c < 0, d > 0)
money supply:
M*S = k1Y + k2r
(k1 > 0, k2 < 0, M*S > 0)
Show that this system can be written as Ax = b, where
1 21 21 0
2a 1 0 0
A5
0 0 1 2c
k1 0 0 k2
Y
C
x5
I
r
b5
G*
b
d
M*s
Use Cramer’s rule to show that
r5
M*(1
2 a) 2 k1(b 1 d 1 G*)
s
k2(1 2 a) 1 ck1
Write down the government expenditure multiplier for r and deduce that the interest
rate, r, increases as government expenditure, G*, increases.
6. The equations defining a model of two trading nations are given by
Y1 = C1 + I *1 + X1 − M1
Y2 = C2 + I *2 + X2 − M2
C1 = 0.6Y1 + 50
C2 = 0.8Y2 + 80
M1 = 0.2Y1
M2 = 0.1Y2
If I *2 = 70, find the value of I *1 if the balance of payments is zero.
[Hint: construct a system of three equations for the three unknowns, Y1, Y2 and I *.]
1
7. The equations defining a general model of two trading countries are given by
Y1 = C1 + I *1 + X1 − M1
Y2 = C2 + I *2 + X2 − M2
C1 = a1Y1 + b1
C2 = a2Y2 + b2
M1 = m1Y1
M2 = m2Y2
where 0 < ai < 1, bi > 0 and 0 < mi < 1 (i = 1, 2). Express this system in matrix form
and use Cramer’s rule to solve this system for Y1. Write down the multiplier for Y1 due
to changes in I *2 and hence give a general description of the effect on the national
income of one country due to a change in investment in the other.
Formal mathematics
In Chapter 3 we introduced the sigma notation as a convenient shorthand for the summation
of series. This notation has an obvious use in matrices where we naturally identify the ele­
ments of a matrix using a double subscript notation. In particular it is possible to formally
define the operation of matrix multiplication. To motivate this, consider forming the matrix
product C = AB where A and B are general 2 × 3 and 3 × 3 matrices with elements, aij and
bij, respectively:
c11 c12
c21 c22
c13
a11
5
c23
a21
a12
a22
a13
a23
b11 b12
b21 b22
b31 b32
b13
b23
b33
If we equate entries in the first row, first column, we get:
c11 = a11b11 + a12b21 + a13b31
3
In sigma notation this could be written as c11 5 o a1k bk1
k51
Similarly, for the second row, third column, we get:
c23 = a21b13 + a22b23 + a23b33
3
which can be written as c23 5 o a2k bk3.
k51
In general, if A is an m × s matrix and B is an s × n matrix, then C = AB is defined to be an m × n
matrix with
s
cij 5 ai1 b1j 1 a i2 b2j 1 a i3 b3j 1 . . . 1 a is bsj 5 o aik bkj
k51
Multiple choice questions
Question 1
If A, B, C and D are 3 × 4, 2 × 3, 4 × 2 and 3 × 3 matrices, which one of the following
matrix calculations is possible?
A. (AC)−1
B. DA + BD
C. D−1B
D. 5CB − 4D−1
E. BD + (AC)T
Question 2
Find the inverse of the matrix
A.
25 3
2 21
B.
1 23
22 5
C.
21 23
22 25
D.
21 3
2 25
E.
21 2
3 25
5 3
.
2 1
Question 3
For what value of a does the following matrix fail to possess an inverse?
21
22
0
A. 7/2
B. 3/2
C. 17/2
D. 23/2
E. 1/2
2 1
3 a
4 6
578 Chapter 7 Matrices
Question 4
3 3 21
0
0 2 23
Find the inverse of 5 4
A.
12 27 23
9 25
215
6 3
210
B.
12 215 210
27
9
6
3
23 25
C.
12 27 24
9 5
215
6 3
210
D.
212
7 4
15 29 25
10 26 23
E.
212
7
4
15 10
29 26
25 23
Question 5
The IS-LM model for a two-sector economy is given by
0.25Y + 10r − 1040 = 0
0.10Y − 36r − 256 = 0
where Y and r denote the equilibrium values of national income and interest rate,
respectively.
Y
If these equations are written in the form Ax = b where x =
, work out the inverse
r
of A and deduce the equilibrium value of r.
A. A−1 =
210
36
, r = 5.36
20.001 0.025
B. A−1 =
3.6
1
, r = 2.8
0.01 20.025
C. A−1 =
3.6
1
,r=4
0.01 20.025
D. A−1 =
21
3.6
, r = 7.44
0.001 0.025
E. A−1 =
21
3.6
, r = 3.4
0.001 0.025
Multiple choice questions 579
Question 6
1 23
1 and B = 3 4
0 21
0 23
If A = 2
A.
11 28
22 26
B.
11 28
22 27
C.
11 8
22 27
D.
9 28
22 27
E.
11 28
1 27
1
, work out BA.
2
Question 7
If A =
3
2 and B = [1 0 −2 3], work out BTAT.
0
21
3 2
0 0
A.
26 24
9 6
0 21
0 0
0 2
0 23
B. [3 0 0 −3]
C. [0]
3
2
D.
0
21
0 26 9
0 24 6
0 0 0
0 2 23
3
0
E.
0
23
Question 8
In a macroeconomic model, Ax = b where
1 21 0
0 1 2a
A=
21 0 1
2t 0 0
0
0
,x=
1
1
Use Cramer’s rule to find C.
Y
I* 1 G*
C
b
and det(A) = 1 − a + at.
,b=
Yd
0
T
T*
580 Chapter 7 Matrices
A.
b 2 aT * 2 a(I* 1 G*)(1 2 t)
1 2 a 1 at
B.
b 2 aT* 2 a(I* 1 G*)(t 2 1)
1 2 a 1 at
C.
b 1 aT * 1 a(I * 1 G*)(t 2 1)
1 2 a 1 at
D.
b 2 aT* 2 at(I* 1 G*)
1 2 a 1 at
E.
(b 2 aT*)t 2 a(I* 1 G*)
1 2 a 1 at
Question 9
Which one of these 2 × 2 matrices is not an inverse of any other matrix in the list?
A.
4 5
5 6
B.
25 6
4 25
C.
25 26
24 25
D.
4 25
25 6
E.
26 5
5 24
Question 10
If A =
p q
a b
and B =
work out ATB
r s
c d
A.
ap 1 bs aq 1 br
cp 1 ds cq 1 dr
B.
ap 1 br aq 1 bs
cp 1 dr cq 1 ds
C.
ap 1 cr aq 1 cs
bp 1 dr bq 1 ds
D.
ar 1 cp aq 1 cs
br 1 dp bq 1 ds
E.
ap 1 cr as 1 cq
bp 1 dr bs 1 dq
Examination questions
Question 1
If A =
1 22
6 23
and B =
, work out, where possible:
0 4
24 2
(a) AT + 2B
(b) AB
(c) A−1
(d) B−1A
Question 2
The demand and supply functions of a good are given by
P + 4QD = 61
3P − QS = 14
Find the equilibrium price and quantity using
(a) the inverse matrix method;
(b) Cramer’s rule.
Question 3
3 21
A=
24 5
1 6
3 0
0 6
4 23
,B=
,C=
22 4
2 2
2 5
21 22
(a) Work out each of the following, where possible:
(i) A + C
(ii) 2A − CT
(iii) AB
(iv) BA
(b) Solve each of the following equations for X:
(i) 2X + AT = C
(ii) BX = A
Question 4
(a) The demand and supply functions of two interdependent commodities are given by
QD1 = 40 − 3P1 −P2
QD2 = 30 − 4P1 − 0.5P2
QS1 = −6 + 5P1
QS2 = −5 + 2P2
Show that the equilibrium prices satisfy
8
1
4 2.5
P1
46
5
P2
35
582 Chapter 7 Matrices
Solve this system by calculating the inverse of the coefficient matrix.
(b) Use Cramer’s rule to find the value of y in the system:
3 5 4
0 2 7
1 3 22
x
212
y 5
0
220
z
Question 5
Airme and Blight are the only two airlines allowed to operate on the same route. The
market share of regular business travellers changes from month to month. Airme retains
four-fifths of its customers whilst Blight retains only three-quarters of its customers.
(a) If At and Bt, denote the number of customers who fly Airme and Blight in month t
explain why
At
0.8 0.25
5
Bt
0.2 0.75
At21
Bt21
(b) If the number of customers who fly Airme and Blight in the first month are 10 000 and
16 000, respectively use matrix multiplication to work out the number of people who
fly Airme
(i) one month later;
(ii) three months later.
(c) If the number of customers flying on Airme and Blight in any month had been 7300 and
5400, respectively, work out how many people flew by each airline the previous month.
Question 6
(a) Find the inverse of the following matrix, in terms of a.
4 22 21
5 1
22
1 a
24
For what value of a is this matrix singular?
(b) Consider the three-commodity market model defined by:
QS1 = 2P1 − 7;
QS2 = 4P2 − 4;
QS3 = 2P3 − 3;
QD1 = −2P1 + 2P2 + P3 + 16
QD2 = 2P1 − P2 − P3 + 8
QD3 = 4P1 − P2 − 4P3 + 4
By making use of your answer to part (a), or otherwise, find the equilibrium prices.
Question 7
The supply and demand equations of two interdependent commodities are
QS1 = 4(P1 − t) − 5,
QS2 = P2 −3, QD1 = 27 − 3P1 + P2
QD2 = 25 + 2P1 − 3P2
where t is the unit tax imposed on good 1.
Examination questions 583
Show that the market equilibrium prices satisfy a matrix equation of the form Ax = b
where x =
P1
.
P2
Find the inverse of A and hence solve this system to find P1 and P2.
Write down the taxation multiplier for P2.
Question 8
Consider the following macroeconomic model:
Y = C + I* + G* + X* − M* (equilibrium of national income)
(consumption; 0 < a < 1; b > 0)
C = aYd + b
T = tY
(taxation; 0 < t < 1)
where I*, G*, X*, M* and Yd denote investment, government expenditure, exports,
imports and disposable income, respectively.
(a) Show that this system can be expressed as Ax = b where
1 21
A = 2a 1
0
2t
0
I 1 G* 1 X * 2 M *
Y
a ,b=
and x = C .
b
1
0
T
(b) Show that the determinant of A is 1 − a(1 − t)
(c) By using Cramer’s rule, or otherwise, show that Y 5
b 1 I * 1 G* 1X * 2 M *
.
1 2 a(1 2 t)
(d) Write down the autonomous investment multiplier for Y and deduce that the change
in national income exceeds any increase in investment.
(e) Work out the marginal propensity to consume multiplier for Y and hence state the
direction of change in Y due to an increase in a.
Question 9
The trace of a 2 × 2 matrix A =
(a) Work out the trace of
a b
is defined to be tr(A) = a + d.
c d
7 4
22 25
(b) Prove that if A and B are any 2 × 2 matrices, then tr(αA + bB) = a tr(A) + b tr(B) for
any numbers a and b.
(c) Decide whether following results are true or false for general 2 × 2 matrices. If the
result is true, prove it, and if the result is false, provide a counter-example.
(i) tr(A−1) = (tr(A))−1
(ii) tr(AB) = tr(BA)
584  
Chapter 8
Linear Programming
Several methods were described in Chapter 5 for optimising functions of two variables subject
to constraints. In economics not all relationships between variables are represented by equations, and we now consider the case when the constraints are given by inequalities. Provided the
function to be optimised is linear and the inequalities are all linear, the problem is said to be one
of linear programming. For simplicity we concentrate on problems involving just two unknowns
and describe a graphical method of solution.
There are two sections, which should be read in the order that they appear. Section 8.1
describes the basic mathematical techniques and considers special cases when problems have
either no solution or infinitely many solutions. Section 8.2 shows how an economic problem,
initially given in words, can be expressed as a linear programming problem and hence solved.
The material in this chapter can be read at any stage, since it requires only an understanding of
how to sketch a straight line on graph paper.
Section 8.1
Graphical solution of linear
programming problems
Objectives
At the end of this section you should be able to:
●
Identify the region defined by a linear inequality.
●
Sketch the feasible region defined by simultaneous linear inequalities.
●
Solve linear programming problems graphically.
●
Appreciate that a linear programming problem may have infinitely many solutions.
●
Appreciate that a linear programming problem may have no finite solution.
In this and the following section we show you how to set up and solve linear programming
problems. This process falls naturally into two separate phases. The first phase concerns
problem formulation; a problem, initially given in words, is expressed in mathematical symbols. The second phase involves the actual solution of such a problem. Experience indicates
that students usually find the first phase the more difficult. For this reason, we postpone
consider­ation of problem formulation until Section 8.2 and begin by investigating techniques
for their mathematical solution.
Advice
You may like to glance at one or two of the examples given in Section 8.2 now to get a feel for the
type of problem that can be solved using these techniques.
Before you can consider linear programming it is essential that you know how to sketch
linear inequalities. In Section 1.1 we discovered that a linear equation of the form
dx + ey = f
can be represented by a straight line on graph paper. We can give a similar graphical interpretation for linear inequalities involving two variables when the equals sign is replaced by one of
<
≤
>
≥
(less than)
(less than or equal to)
(greater than)
(greater than or equal to)
To illustrate this, consider the simple inequality
y≥x
Section 8.1 Graphical solution of linear programming problems 587
y
P
R
Q
x
y5 x
Figure 8.1
We would like to identify those points with coordinates (x, y) for which this inequality is
true. Clearly, this has something to do with the straight line
y=x
sketched in Figure 8.1.
If a point P lies above the line, then the y coordinate is greater than the x coordinate, so
that
y>x
Similarly, if a point Q lies below the line, then the y coordinate is less than the x coordinate, so that
y<x
Of course, the coordinates of a point R which actually lies on the line satisfy
y=x
Hence we see that the inequality
y≥x
holds for any point that lies on or above the line y = x.
It is useful to be able to indicate this region pictorially. We do this by shading one half of
the coordinate plane. There are actually two schools of thought here. Some people like to
shade the region containing the points for which the inequality is true. Others prefer to shade
the region for which it is false. In this text we adopt the latter approach and always shade the
region that we are not interested in, as shown in Figure 8.2. This may seem a strange choice,
but the reason for making it will soon become apparent.
In general, to sketch an inequality of the form
dx + ey < f
dx + ey ≤ f
dx + ey > f
dx + ey ≥ f
588 Chapter 8 Linear Programming
y
y$ x
x
Figure 8.2
we first sketch the corresponding line
dx + ey = f
and then decide which side of the line to deal with. An easy way of doing this is to pick a
‘test point’, (x, y). It does not matter what point is chosen, provided it does not actually lie
on the line itself. The numbers x and y are then substituted into the original inequality. If the
inequality is satisfied, then the side containing the test point is the region of interest. If not,
then we go for the region on the other side of the line.
For the region
2x + y < 4
we first sketch the line
2x + y = 4
When x = 0, we get
y=4
When y = 0, we get
2x = 4
and so x = 4/2 = 2.
The line passes through (0, 4) and (2, 0) and is shown in Figure 8.3. For a test point let us
take (3, 2), which lies above the line. Substituting x = 3 and y = 2 into the expression 2x + y
gives
2(3) + 2 = 8
This is not less than 4, so the test point does not satisfy the inequality. It follows that
the region of interest lies below the line. This is illustrated in Figure 8.4. In this example
the symbol < is used rather than ≤. Hence the points on the line itself are not included in the
region of interest. We have chosen to indicate this by using a broken line for the boundary.
Section 8.1 Graphical solution of linear programming problems 589
y
5
4
3
2
(3, 2)
1
25 24 23 22 21
21
1
2
3
4
22
5
x
2x 1 y 5 4
23
24
25
Figure 8.3
y
5
4
3
2x 1 y , 4
2
1
25 24 23 22 21
21
1
2
3
4
5
x
22
23
24
25
Figure 8.4
Practice Problem
1. Sketch the straight line
−x + 3y = 6
on graph paper. By considering the test point (1, 4), indicate the region
−x + 3y > 6
590 Chapter 8 Linear Programming
y
x$0
y$0
x
Figure 8.5
We now consider the region defined by simultaneous linear inequalities. This is known
as a feasible region. It consists of those points (x, y) which satisfy several inequalities at the
same time. We find it by sketching the regions defined by each inequality in turn. The feasible
region is then the unshaded part of the plane corresponding to the intersection of all of the
individual regions.
To illustrate this, consider the feasible region defined by
x + 2y ≤ 12
−x + y ≤ 3
x≥0
y≥0
In this problem the easiest inequalities to handle are the last two. These merely indicate that
x and y are non-negative, and so we need only consider points in the top right-hand quadrant
of the plane, as shown in Figure 8.5.
For the inequality
x + 2y ≤ 12
we need to sketch the line
x + 2y = 12
When x = 0, we get
2y = 12
and so y = 12/2 = 6.
When y = 0, we get
x = 12
The line passes through (0, 6) and (12, 0).
For a test point let us take (0, 0), since such a choice minimises the amount of arithmetic
that we have to do. Substituting x = 0 and y = 0 into the inequality gives
0 + 2(0) ≤ 12
which is obviously true. Now the region containing the origin lies below the line, so we shade
the region that lies above it. This is indicated in Figure 8.6.
Section 8.1 Graphical solution of linear programming problems 591
y
8
7
6
5
4
3
x 1 2y # 12
2
1
1
2
3
4
5
6
7
8
9
10
11
12
13
x
Figure 8.6
For the inequality
−x + y ≤ 3
we need to sketch the line
−x + y = 3
When x = 0, we get
y=3
When y = 0, we get
−x = 3
and so x = 3/(−1) = −3.
The line passes through (0, 3) and (−3, 0). Unfortunately, the second point does not lie
on the diagram as we have drawn it. At this stage we can either redraw the x axis to include
−3, or we can try finding another point on the line which does fit on the graph. For example,
putting x = 5 gives
−5 + y = 3
so y = 3 + 5 = 8. Hence the line passes through (5, 8), which can now be plotted along with (0, 3)
to sketch the line. At the test point (0, 0) the inequality reads
−0 + 0 ≤ 3
which is obviously true. We are therefore interested in the region below the line, since
this contains the origin. As usual, we indicate this by shading the region on the other side.
The complete picture is shown in Figure 8.7.
Points (x, y) which satisfy all four inequalities must lie in the unshaded ‘hole’ in the
middle. Incidentally, this explains why we did not adopt the convention of shading the region
of interest. Had we done so, our task would have been to identify the most heavily shaded
part of the diagram, which is not so easy.
592 Chapter 8 Linear Programming
y
9
8
(5, 8)
7
6
5
4
3
feasible
region
2
1
1
2
3
4
5
6
7
8
9
10
11
12
13
x
Figure 8.7
Practice Problem
2. Sketch the feasible region
x + 2y ≤ 10
3x + y ≤ 10
x≥0
y≥0
We are now in a position to explain exactly what we mean by a linear programming problem
and how such a problem can be solved graphically. We actually intend to describe two
slightly different methods of solution. One of these is fairly sophisticated and difficult to use,
while the other is more straightforward. The justification for bothering with the ‘harder’
method is that it provides the motivation for the ‘easier’ method. It also helps us to handle
one or two trickier problems that sometimes arise. We shall introduce both methods by concentrating on a specific example:
Solve the linear programming problem:
Minimise
−2x + y
subject to the constraints
x + 2y ≤ 12
−x + y ≤ 3
x≥0
y≥0
In general, three ingredients make up a linear programming problem. First, there are several
unknowns to be determined. In this example there are just two unknowns, x and y. Secondly,
there is a mathematical expression of the form
ax + by
Section 8.1 Graphical solution of linear programming problems 593
which we want to either maximise or minimise. Such an expression is called an objective function.
In this example, a = −2, b = 1 and the problem is one of minimisation. Finally, the unknowns
x and y are subject to a collection of linear inequalities. Quite often (but not always) two of
the inequalities are x ≥ 0 and y ≥ 0. These are referred to as non-negativity constraints. In this
example there are a total of four constraints including the non-negativity constraints.
Geometrically, points (x, y) which satisfy simultaneous linear inequalities define a feasible
region in the coordinate plane. In fact, for this particular problem, the feasible region has
already been sketched in Figure 8.7.
The problem now is to try to identify that point inside the feasible region which minimises
the value of the objective function. One naïve way of doing this might be to use trial and error:
that is, we could evaluate the objective function at every point within the region and choose
the point which produces the smallest value. For instance, (1, 1) lies in the region, and when
the values x = 1 and y = 1 are substituted into
−2x + y
we get
−2(1) + 1 = −1
Similarly we might try (3.4, 2.1), which produces
−2(3.4) + 2.1 = −4.7
which is an improvement, since −4.7 < −1.
The drawback of this approach is that there are infinitely many points inside the region, so
it is going to take a very long time before we can be certain of the solution! A more systematic
approach is to superimpose, on top of the feasible region, the family of straight lines,
−2x + y = c
for various values of the constant c. Looking back at the objective function, you will notice
that the number c is precisely the thing that we want to minimise. That such an equation
represents a straight line should come as no surprise to you by now. Indeed, we know from
the rearrangement
y = 2x + c
that the line has a slope of 2 with a y intercept of c. Consequently, all of these lines are parallel
to each other, their precise location being determined from the number c.
Now, when y = 0, the equation reads
0 = 2x + c
and so has solution x = −c/2. Hence the line passes through the point (−c/2, 0). A selection
of lines is sketched in Figure 8.8 for values of c in the range 0 to −24. These have been
sketched using the information that they pass through (−c/2, 0) and have a slope of 2.
Note that as c decreases from 0 to −24, the lines sweep across the feasible region from left to
right. Also, once c goes below −24, the lines no longer intersect this region. The minimum
value of c (which, you may remember, is just the value of the objective function) is therefore
−24. Moreover, when c = −24, the line
−2x + y = c
intersects the feasible region in exactly one point, namely (12, 0). This, then, must be the
solution of our problem. The point (12, 0) lies in the feasible region as required and because
it also lies on the line
−2x + y = −24
594 Chapter 8 Linear Programming
c50
c decreases
c 5 24
c 5 28
c 5 212
c 5 216
c 5 220
c 5 224
y
6
(2, 5)
5
4
(0, 3)
3
2
1
(0, 0)
1
2
3
4
5
6
7
8
9
10
11
12
x
(12, 0)
Figure 8.8
we know that the corresponding value of the objective function is −24, which is the minimum
value. Other points in the feasible region also lie on lines
−2x + y = c
but with larger values of c.
Practice Problem
3. Consider the linear programming problem
Maximise
−x + y
subject to the constraints
3x + 4y ≤ 12
x≥0
y≥0
(a) Sketch the feasible region.
(b) Sketch, on the same diagram, the five lines
y=x+c
for c = −4, −2, 0, 1 and 3.
[Hint: lines of the form y = x + c have a slope of 1 and pass through the points (0, c)
and (−c, 0).]
(c) Use your answers to part (b) to solve the given linear programming problem.
In the previous problem, and in Practice Problem 3, the optimal value of the objective function
is attained at one of the corners of the feasible region. This is not simply a coincidence. It can
be shown that the solution of any linear programming problem always occurs at one of the
corners. Consequently, the trial-and-error approach suggested earlier is not so naïve after all.
The only possible candidates for the answer are the corners, and so only a finite number of
points need ever be examined. This method may be summarised:
Section 8.1 Graphical solution of linear programming problems 595
Step 1
Sketch the feasible region.
Step 2
Identify the corners of the feasible region and find their coordinates.
Step 3
Evaluate the objective function at the corners and choose the one which has the maximum or
minimum value.
Returning to the previous example, we work as follows:
Step 1
The feasible region has already been sketched in Figure 8.7.
Step 2
There are four corners with coordinates (0, 0), (0, 3), (2, 5) and (12, 0).
Step 3
Corner
Objective function
(0, 0)
−2(0) + 0 = 0
(0, 3)
−2(0) + 3 = 3
(2, 5)
−2(2) + 5 = 1
(12, 0)
−2(12) + 0 = −24
From this we see that the minimum occurs at (12, 0), at which the objective function is
−24. Incidentally, if we also require the maximum, then this can be deduced without further
effort. From the table the maximum is 3, which occurs at (0, 3).
Example
Solve the linear programming problem
Maximise
5x + 3y
subject to
2x + 4y ≤ 8
x≥0
y≥0
Solution
Step 1
The non-negativity constraints x ≥ 0 and y ≥ 0 indicate that the region is bounded by the
coordinate axes in the positive quadrant.
The line 2x + 4y = 8 passes through (0, 2) and (4, 0). Also, at the test point (0, 0), the inequality
2x + 4y ≤ 8
reads
0≤8
which is true. We are therefore interested in the region below the line, since this region
contains the test point, (0, 0). The feasible region is sketched in Figure 8.9.
➜
596 Chapter 8 Linear Programming
y
3
(0, 2)
2
1
feasible
region
(4, 0)
(0, 0)
1
2
3
4
5
x
Figure 8.9
Step 2
The feasible region is a triangle with three corners, (0, 0), (0, 2) and (4, 0).
Step 3
Corner
Objective function
(0, 0)
5(0) + 3(0) = 0
(0, 2)
5(0) + 3(2) = 6
(4, 0)
5(4) + 3(0) = 20
The maximum value of the objective function is 20, which occurs when x = 4 and y = 0.
Practice Problems
4. Solve the linear programming problem
Minimise
x−y
subject to
2x + y ≤ 2
x≥0
y≥0
5. Solve the linear programming problem
Maximise
3x + 5y
subject to
x + 2y ≤ 10
3x + y ≤ 10
x≥0
y≥0
[Hint: you might find your answer to Practice Problem 2 useful.]
Section 8.1 Graphical solution of linear programming problems 597
In Section 1.4 we showed you how to solve a system of simultaneous linear equations.
We discovered that such a system does not always have a unique solution. It is possible for a
problem to have either no solution or infinitely many solutions. An analogous situation arises
in linear programming. We conclude this section by considering two examples that illustrate
these special cases.
Example
Solve the linear programming problem
Maximise
x + 2y
subject to
2x + 4y ≤ 8
x≥0
y≥0
Solution
Step 1
The feasible region is identical to the one sketched in Figure 8.9 for the previous worked
example.
Step 2
As before, the feasible region has three corners, (0, 0), (0, 2) and (4, 0).
Step 3
Corner
Objective function
(0, 0)
0 + 2(0) = 0
(0, 2)
0 + 2(2) = 4
(4, 0)
4 + 2(0) = 4
This time, however, the maximum value is 4, which actually occurs at two corners, (0, 2)
and (4, 0). This shows that the problem does not have a unique solution. To explain what
is going on here, we return to the method introduced at the beginning of this section. We
superimpose the family of lines obtained by setting the objective function equal to some
constant, c. The parallel lines
x + 2y = c
pass through the points (0, c/2) and (c, 0).
A selection of lines is sketched in Figure 8.10 for values of c between 0 and 4. These
particular values are chosen since they produce lines that cross the feasible region. As c
increases, the lines sweep across the region from left to right. Moreover, when c goes
above 4, the lines no longer intersect the region. The maximum value that c (that is, the
objective function) can take is therefore 4. However, instead of the line
x + 2y = 4
intersecting the region at only one point, it intersects along a whole line segment of points.
Any point on the line joining the two corners (0, 2) and (4, 0) will be a solution. This follows
because any point on this line segment lies in the feasible region, and the corresponding
value of the objective function on this line is 4, which is the maximum value.
➜
598 Chapter 8 Linear Programming
y
2
1
1
2
3
c5 1
4
c5 2
c5 3
x
c5 4
c increases
Figure 8.10
This example suggests the general result. If, in step 3, the maximum (or minimum) occurs
at two corners, then the problem has infinitely many solutions. Any point on the line segment
joining these corners, including the two corners themselves, is also a solution.
Example
Solve the linear programming problem
Maximise
3x + 2y
subject to
x + 4y ≥ 8
x+y≥5
2x + y ≥ 6
x≥0
y≥0
What can you say about the solution if this problem is one of minimisation rather than
maximisation?
Solution
Step 1
As usual, the non-negativity constraints indicate that we need only consider the positive
quadrant.
The line x + 4y = 8 passes through (0, 2) and (8, 0).
The line x + y = 5 passes through (0, 5) and (5, 0).
The line 2x + y = 6 passes through (0, 6) and (3, 0).
Section 8.1 Graphical solution of linear programming problems 599
y
unbounded
feasible
region
7
6
(0, 6)
5
(1, 4)
4
3
2
(4, 1)
1
(8, 0)
1
2
3
4
5
6
7
8
9
x
Figure 8.11
Also, the test point (0, 0) does not satisfy any of the corresponding constraints because
the three inequality signs are all ‘≥’. We are therefore interested in the region above all of
these lines, as shown in Figure 8.11.
Step 2
The feasible region has four corners, (0, 6), (1, 4), (4, 1) and (8, 0).
Step 3
Corner
Objective function
(0, 6)
(1, 4)
(4, 1)
3(0) + 2(6) = 12
3(1) + 2(4) = 11
3(4) + 2(1) = 14
(8, 0)
3(8) + 2(0) = 24
From the table, the minimum and maximum values of the objective function are 11 and
24, which occur at (1, 4) and (8, 0), respectively. However, we do have a slightly unusual
situation in that the feasible region is not enclosed on all sides. We describe this by saying
that the feasible region is unbounded. It is open at the top and, strictly speaking, it does
not make sense to talk about the corners of such a region. Are we therefore justified
in applying the ‘easy’ method in this case? To answer this question we superimpose the
family of lines
3x + 2y = c
representing the objective function, as shown in Figure 8.12.
When c = 11, the line intersects the region at only one point (1, 4). However, as c
increases from this value, the lines sweep across the feasible region and never leave it, no
matter how large c becomes. Consequently, if the problem is one of maximisation, we
conclude that it does not have a finite solution. We can substitute huge values of x and y
into 3x + 2y and get an ever-increasing result. On the other hand, if the problem is one of
minimisation, then it does have a solution at the corner (1, 4). This, of course, is the
answer obtained previously using the ‘easy’ method.
➜
600 Chapter 8 Linear Programming
y
14
13
12
11
10
9
c increases
8
7
6
5
4
3
2
1
1
2
3
4
5
c 5 11
6
7
8
c 5 18
9 10 11 12 13 14
x
c 5 24
Figure 8.12
This example shows that a linear programming problem may not have a finite solution
when the feasible region is unbounded. However, when a solution does exist, it may be found
simply by inspecting the corners in the normal way. In practice, linear programming problems
arise from realistic economic situations. We would therefore expect the problem to possess a
sensible (that is, finite) answer, and so the difficulty of the non-existence of a solution rarely
occurs.
Key Terms
Feasible region The set of points which satisfy all of the constraints in a linear programming
problem.
Non-negativity constraints The constraints x ≥ 0, y ≥ 0, etc.
Objective function The function that is optimised in a linear programming problem.
Unbounded region A feasible region that is not completely enclosed by a polygon. The associated
linear programming problem may not have a finite solution.
Section 8.1 Graphical solution of linear programming problems 601
Exercise 8.1
1. Which of the following points satisfy the inequality
2x − 3y > −5?
(1, 1), (−1, 1), (1, −1), (−1, −1), (−2, 1), (2, −1), (−1, 2) and (−2, −1)
2. How many points with integer coordinates lie in the feasible region defined by
3x + 4y ≤ 12,
x ≥ 0 and
y ≥ 1?
3. Sketch the feasible regions defined by the following sets of inequalities:
(a) 5x + 3y ≤ 30 (b) 2x + 5y ≤ 20 (c) x − 2y ≤ 3
7x + 2y ≤ 28
x ≥ 0 
y ≥ 0 
x+y≤5
x≥0
y≥0
x−y≤4
x≥1
y≥0
4. Use your answers to Question 3 to solve the following linear programming problems:
(a) Maximise
4x + 9y
subject to
5x + 3y ≤ 30
7x + 2y ≤ 28
x≥0
y≥0
(b) Maximise
3x + 6y
subject to
2x + 5y ≤ 20
x+y≤5
x≥0
y≥0
(c) Minimise
x+y
subject to
x − 2y ≤ 3
x−y≤4
x≥1
y≥0
5. What can you say about the solution to Question 4(c) if the problem is one of maximisa-
tion rather than minimisation? Explain your answer by superimposing the family of
lines
x+y=c
on the feasible region.
6. Find, if possible, the minimum value of the objective function 3x − 4y subject to the
constraints,
−2x + y ≤ 12, x − y ≤ 2,
x≥0
and y ≥ 0
➜
602 Chapter 8 Linear Programming
7. What can you say about the solution of the linear programming problem specified in
Question 6, if the
(a) objective function is to be maximised instead of minimised?
(b) second constraint is changed to x + y ≤ 2 and the problem is one of minimisation?
(c) second constraint is changed to 3x − 4y ≤ 24, and the problem is one of maximisation?
Exercise 8.1*
1. The point (x, 3) satisfies the inequality, −5x − 2y ≤ 13. Find the smallest possible value
of x.
2. The following five inequalities define a feasible region. Which one of these could be
removed from the list without changing the region?
A: −x + y ≤ 10
B:
x + y ≤ 20
C: x − 2y ≥ −8
D:
x≥0
E:
y≥0
3. Solve the following linear programming problems:
(a) Maximise
2x + 3y (b) Maximise
−8x + 4y
subject to
subject to
2x + y ≤ 8 x − y ≤ 2
x+y≤6
2x − y ≥ −3
x + 2y ≤ 10 x − y ≥ −4
x ≥ 0
x≥0
y ≥ 0
y≥0
4. Explain why each of the following problems fails to possess a solution:
(a) Maximise
x + y (b) Maximise x + y
subject to
subject to
y≥2
2x − y ≥ −1
x≤2
x − 2y ≤ 2
x − y ≤ 1
x≥0
x ≥ 0
y≥0
y ≥ 0
5. Solve the linear programming problem
Maximise 6x + 2y
subject to
x−y≥0
3x + y ≥ 8
x≥0
y≥0
Section 8.1 Graphical solution of linear programming problems 603
6. Show that the linear programming problem given in Question 3(a) can be expressed in
matrix notation as
Maximise cTx
subject to
Ax ≤ b
x≥0
where c, x, and 0 are 2 × 1 matrices, A is a 3 × 2 matrix and b is a 3 × 1 matrix, which
should be stated.
7. (a) Consider the linear programming problem
Minimise x + y
subject to
2x + y ≥ 16
2x + 3y ≥ 24
−x + y ≤ 12
x≥0
y≥0
(i) Sketch the feasible region and find the exact coordinates of the corners.
(ii) Tabulate the values of the objective function at each of the corners and hence
state the coordinates of the optimal point.
(iii) One of the five constraints is not needed for drawing the feasible region. State
the redundant constraint.
(b) What can you say about the solution if the problem is changed to one of
(i) maximising x + y
(ii) minimising 2x + y
and is subject to the same constraints as part (a)? Give reasons for your answers.
(c) The objective function in part (a) is changed to
ax + 2y
where a is a positive constant. The problem remains one of minimisation subject to
the same constraints as before.
Find the largest value of a for which the linear programming problem has solution
x = 12, y = 0. Explain your reasoning carefully.
Section 8.2
Applications of linear programming
Objectives
At the end of this section you should be able to:
●
Identify the decision variables in a linear programming problem.
●
Find an expression for the objective function and decide whether it should be maximised or
minimised.
●
Write down all of the constraints in the problem specification.
●
Solve linear programming problems expressed in words, remembering to confirm that the answer
makes sense.
●
Work out and interpret shadow prices.
The impression possibly given so far is that linear programming is a mathematical technique
designed to solve rather abstract problems. This is misleading, since linear programming
problems do arise from concrete situations.
In fact, linear programming is a vital tool for business managers who need to allocate finite
resources such as labour, machine time or raw materials to maximise profits or minimise
costs. The word ‘programming’ is used in the sense of scheduling or management planning.
In this section we develop an important skill that can loosely be called problem formulation.
Here we start with information, perhaps only vaguely given in words, and try to express it
using the more precise language of mathematics. If the linear programming problem involves
just two variables we can then use the graphical technique described in Section 8.1 to solve
it. In the real world a problem could involve many more variables. Under these circumstances
we would use a specialist computer package based on the simplex method to solve it.
However, in order to describe problem formulation, we begin with an example involving just
two variables.
A craft studio makes glass bowls and plates, and it can sell all of the products that it makes
each week. Glassware is made in two stages. In the first stage molten glass is taken from a
furnace and placed at the end of a blowpipe. Skilled glassblowers then mould the glass into
the desired shape. During the second stage the glass is allowed to cool in a controlled way
inside a second furnace (called an annealer) which helps to reduce stress and prevent the glass
from cracking. The studio employs two glassblowers who each work a 35-hour week, and
the total time available each week for annealing is 130 hours. The main raw material used in
glassmaking is silica sand, and the studio orders 45 kg each week. A glass bowl and plate each
require 1 kg of sand. It takes a glassblower two hours to make a bowl and one hour to make
a plate. The time needed to cool a bowl in an annealer is four hours and a plate requires one
hour. The profit made from selling a bowl and a plate is $150 and $100, respectively. Of
course, if there were no restrictions on either time or materials, the studio would be able to
maximise weekly profit by making only glass bowls. However, with limited resources, we use
linear programming to decide on the best combination.
Section 8.2 Applications of linear programming 605
As mentioned in Section 8.1, there are three things constituting a linear programming
problem: unknowns, an objective function which needs maximising or minimising, and some
constraints. We consider each of these in turn.
The studio needs to decide how many bowls and plates to make each week. These are
therefore the unknowns (called the decision variables) of this problem, and we denote these by
the letters, x and y: that is, we let
●
x = number of bowls
●
y = number of plates
The object is to maximise profit, so we need to find an expression for this in terms of x and
y. The studio makes a profit of $150 for every bowl and, since it makes x of these, the total
profit made is 150x. Similarly, the profit made from sales of the plates is 100y, so the combined profit made from both bowls and plates is
150x + 100y
This is the objective function that we want to maximise.
We now consider the limited resources available and use the information provided to
obtain the constraints. The total amount of time available each week for glassblowing is
70 hours, and the time needed to make a bowl and a plate is two hours and one hour, respectively. Hence if x bowls and y plates are made, the total time needed is
2x + y
so we require
2x + y ≤ 70
The total amount of time needed for annealing is
4x + y
and since this must not exceed the 130 hours available, we need
4x + y ≤ 130
Similarly, the amount of sand ordered is 45 kg, so
x + y ≤ 45
On the face of it there appears to be no further constraints given in the problem. However,
a moment’s thought should convince you that we are missing two important constraints,
namely the non-negativity constraints
x ≥ 0,
y≥0
Although these are not mentioned explicitly, it is obviously not possible to make a negative
number of goods. The complete statement of the problem may now be stated:
Maximise 150x + 100y (profit)
subject to
2x + y ≤ 70
4x + y ≤ 130
x + y ≤ 45
x≥0
y≥0
(glassblowing time)
(annealing time)
(mass of silica sand)
The problem can now be solved using the method described in Section 8.1.
606 Chapter 8 Linear Programming
y
150
140
130
annealing
120
110
100
90
80
70
glassblowing
60
50
(0, 45)
40
30
sand
20
feasible region
10
0
(0, 0)
(25, 20)
(30, 10)
5
10
15
20
25
30
35
(32.5, 0)
40
45
x
Figure 8.13
Step 1
As usual, the non-negativity constraints indicate that we need only consider points in the
positive quadrant.
The line 2x + y = 70 passes through (0, 70) and (35, 0)
The line 4x + y = 130 passes through (0, 130) and (32.5, 0)
The line x + y = 45 passes through (0, 45) and (45, 0)
Also, using the origin as the test point reveals that the region of interest lies below all three
lines. It is sketched in Figure 8.13.
Step 2
The feasible region has five corners: (0, 0), (0, 45), (25, 20), (30, 10) and (32.5, 0).
Step 3
Corner
Profit
(0, 0)
(0, 45)
(25, 20)
(30, 10)
(32.5, 0)
0
4500
5750
5500
4875
The maximum weekly profit is $5750, which occurs when 25 bowls and 20 plates are made.
The corner giving the optimal solution, (25, 20), is the point of the intersection of the lines
x + y = 45 and 2x + y = 70 which correspond to the constraints arising from sand and glassblowing. Mathematically, for these two resources the inequality ‘≤’ has been replaced by
equality, ‘=’, so the glassblowing time and the quantity of sand are fully utilised. On the other
hand, the line for annealing does not pass through the optimum, so there is some slack in this
constraint. Substituting x = 25 and y = 20 into the expression for the annealing time gives
4x + y = 4 × 25 + 20 = 120
There were 130 hours available, so 10 of these hours are unused.
Section 8.2 Applications of linear programming 607
Advice
It is impossible to give a precise description of problem formulation. Each case has to be considered
on its own merits. However, you might like to try the following general strategy:
(1) Identify the decision variables and label them x and y.
(2) Write down an expression for the objective function in terms of x and y, and decide whether it needs
maximising or minimising.
(3) Write down all constraints on the variables x and y, including the non-negativity constraints.
Practice Problem
1. An electronics firm decides to launch two models of a tablet, TAB1 and TAB2. The cost
of making each device of type TAB1 is $120 and the cost for TAB2 is $160. The firm
recognises that this is a risky venture, so it decides to limit the total weekly production
costs to $4000. Also, due to a shortage of skilled labour, the total number of tablets
that the firm can produce in a week is at most 30. The profit made on each device is
$600 for TAB1 and $700 for TAB2. How should the firm arrange production to maximise profit?
One important task when planning business projects is to decide whether it is worthwhile
buying in extra resources. For the glassmaking problem the studio might consider the option
of increasing the number of hours available for glassblowing or annealing. It might also
investigate what happens to the maximum profit if it buys an extra kilogram of sand. Do
these changes make any difference to the number of bowls and plates that the studio should
make to maximise profit? These questions can be answered by reworking the problem changing one constraint at a time. Most software packages provide this information automatically
so that management can see at a glance how sensitive the optimal solution is to small changes
in resources.
To illustrate this idea let us consider the case when one of the glassblowers works for an
extra hour a week but all other resources are unchanged. The only difference to the problem
specification is that the glassblowing constraint changes from
2x + y ≤ 70
to
2x + y ≤ 71
All that happens is that the line joining (0, 70) and (35, 0) moves to a line joining (0, 71) and
(35.5, 0). The feasible region is barely altered, so the optimal point will still be at the intersection of the lines representing the boundaries of the glassblowing and sand constraints. Rather
than attempt to draw these accurately on a diagram, it is easier to find the new point by solving the simultaneous equations
x + y = 45
2x + y = 71
Subtracting the first equation from the second gives x = 26. If this value is substituted into
either equation, it is easy to see that y = 19. The new profit is worked out to be
150 × 26 + 100 × 19 = $5800
608 Chapter 8 Linear Programming
This compares with a previous profit of $5750, so a 1-unit increase in the number of hours
available for glassblowing results in a $50 increase in profit. This value is called the shadow
price and is similar to the marginal concepts that we have met before. The shadow price
represents the maximum premium that a firm would be willing to pay for 1 extra unit of that
resource. The studio will welcome the extra hour of glassblowing provided it costs less than
$50 to provide it. In this example the shadow price of annealing is obviously zero because we
already know that 10 hours of annealing time is unused, so providing an extra hour for this
would have no impact on the overall profit. This is the case for any resource when there is
slack in the constraint (provided that the change is not so great that the optimum point in the
feasible region moves from one corner to a different corner.)
Practice Problem
2. For the glass studio problem find the coordinates of the optimal corner of the feasible
region when an extra kilogram of sand becomes available, and hence work out the
shadow price of sand.
Example
An insurance company employs full- and part-time staff, who work 40 and 20 hours per
week, respectively. Full-time staff are paid $800 per week and part-time staff $320. In addition, it is company policy that the number of part-time staff should not exceed one-third of
the number of full-time staff.
If the number of worker-hours per week required to deal with the company’s work is
900, how many workers of each type should be employed in order to complete the workload
at minimum cost?
Solution
If the company employs x full-time staff and y part-time staff, then the company would
like to choose x and y to minimise its weekly salary costs. Also, since full- and part-time
staff are paid $800 and $320 per week, respectively, the total wage bill is then
800x + 320y
which is the objective function that needs to be minimised.
Full- and part-time staff work 40 and 20 hours per week, respectively, so the total number of worker-hours available is
40x + 20y
It is required that this is at least 900, so we obtain the constraint
40x + 20y ≥ 900
Section 8.2 Applications of linear programming 609
A further constraint on the company arises from the fact that the number of part-time
staff cannot exceed one-third of the number of full-time staff. This means, for example,
that if the company employs 30 full-time staff, then it is not allowed to employ more than
10 part-time staff because
1/3 × 30 = 10
In general, if x denotes the number of full-time staff, then the number of part-time staff,
y, cannot exceed x/3: that is,
y ≤ x/3
In addition, we have the obvious non-negativity constraints
x≥0
and
y≥0
The complete problem may now be stated:
800x + 320y
Minimise
subject to
40x + 20y ≥ 900
y ≤ x/3
x≥0
y≥0
It can now be solved by applying the method described in Section 8.1.
Step 1
The feasible region can be sketched in the usual way. The line y = x/3 passes through
(0, 0), (3, 1), (6, 2) and so on. Unfortunately, because the origin actually lies on the line,
it is necess­ary to use some other point as a test point. For example, substituting x = 30,
y = 5 into the inequality
y ≤ x/3
gives
5 ≤ (30)/3
This inequality is clearly true, indicating that (30, 5), which lies below the line, is in the
region of interest. The constraint
40x + 20y ≥ 900
is easier to handle. The corresponding line passes through (0, 45) and (221/2, 0), and using
the origin as a test point shows that we need to shade the region below the line. The feasible
region is sketched in Figure 8.14.
Step 2
The feasible region has two corners. One of these is obviously (221/2, 0). However, it is not
possible to write down directly from the diagram the coordinates of the other corner. This
is formed by the intersection of the two lines
y = x/3
40x + 20y = 900
(1)
(2)
➜
610 Chapter 8 Linear Programming
y
50
45
40
35
30
25
20
15
10
feasible
region
(?, ?)
5
1
(22 2 ,
5
10
15
20
25
0)
30
35
40
45
50
x
Figure 8.14
and so we must solve this system algebraically. In this case the easiest thing to do is to
substitute equation (1) into equation (2) to eliminate y immediately. This gives
40x 1
20
x 5 900
3
that is,
140
x 5 900
3
which has the solution
2700 135
5
5 192/7
140
7
x5
Finally, from equation (1),
1
1 135 45
3
x5 3
5
56
3
3
7
7
7
y5
The feasible region therefore has coordinates (192/7, 63/7) and (221/2, 0).
Step 3
Corner
2
Objective function
3
(19 /7, 6 /7)
(221/2, 0)
17 4855/7
18 000
The minimum cost is $17 485.71, which occurs when x = 192/7 and y = 63/7.
It might appear that this is the solution to our original problem. This is certainly mathe­
matically correct, but it cannot possibly be the solution that we are looking for, since it
does not make sense, for example, to employ 2/7 of a worker. We are interested only in
Section 8.2 Applications of linear programming 611
y
7
(21, 7)
2
3
(197 , 6 7 )
(20, 6)
6
(21, 6)
(20, 5)
(21, 5)
5
19
20
21
22
x
Figure 8.15
points whose coordinates are whole numbers. A problem such as this is referred to as an
integer programming problem. We need to find that point (x, y) inside the feasible region
where both x and y are whole numbers which minimises the objective function. A ‘blow-up’
picture of the feasible region near the minimum cost corner is shown in Figure 8.15, and
the following table indicates that the optimal solution is (20, 5).
Corner
Objective function
(20, 5)
(20, 6)
(21, 5)
(21, 6)
17 600
17 920
18 400
18 720
Other points in the neighbourhood with whole-number coordinates are (20, 6), (20, 7),
(21, 5) and so on. However, all of these have either a larger value of x or a larger value of y
(or both) and so must produce a larger total cost. The company should therefore employ
20 full- and 5 part-time staff to minimise its salary bill.
Advice
This example highlights the need to look back at the original problem to make sure that the final
answer makes sense. It is very tempting when solving linear programming problems just to write
down the solution without thinking, neatly underline it and then go on to another problem.
Unfortunately, it is all too easy to make mistakes, both in formulating the problem and in sketching
the feasible region. Spending a few moments checking the validity of your solution may well help you
to discover any blunders that you have made and may suggest possible modifications to the solution
procedure, as in the previous example. We might, therefore, conclude the general strategy with the
following step:
(4) Check that the final answer makes sense as a solution to the original problem.
612 Chapter 8 Linear Programming
Practice Problem
3. An individual spends 95% of earned income on essential goods and services, leaving
only 5% to be spent on luxury goods, which is subdivided between trendy clothes and
visits to the theatre. The cost of each item of clothing is $150 and a trip to the theatre
costs $70. The corresponding utility function is
U = 3x + 7y
where x and y denote the number of trendy clothes and theatre visits per year, respectively. In order to maintain a reasonable appearance throughout the year, it is vital that
at least nine new items of clothing are purchased each year. Given that annual earned
income is $42 000, find the values of x and y which maximise utility.
Key Terms
Decision variable
The unknowns in a linear programming problem which can be controlled.
Integer programming A linear programming problem in which the search for solution is
restricted to points in the feasible region with whole-number coordinates.
Shadow price The change in the optimal value of the objective function due to a 1-unit increase
in one of the available resources.
Exercise 8.2
1. A manufacturer produces two models of racing bike, B and C, each of which must be
pro­cessed through two machine shops. Machine shop 1 is available for 120 hours per
month and machine shop 2 for 180 hours per month. The manufacture of each bike of
type B takes six hours in shop 1 and three hours in shop 2. The corresponding times for
C are 4 and 10 hours, respectively. If the profit is $180 and $220 per bike of type B and C,
respectively, how should the manufacturer arrange production to maximise total profit?
2. A small firm manufactures and sells litre cartons of non-alcoholic cocktails, ‘The
Caribbean’ and ‘Mr Fruity’, which sell for $1 and $1.25, respectively. Each is made by
mixing fresh orange, pineapple and apple juices in different proportions. The Caribbean
consists of one part orange, six parts pineapple and one part apple. Mr Fruity consists
of two parts orange, three parts pineapple and one part apple. The firm can buy up to
300 litres of orange juice, up to 1125 litres of pineapple juice and up to 195 litres of
apple juice each week at a cost of $0.72, $0.64 and $0.48 per litre, respectively.
Find the number of cartons of ‘The Caribbean’ and ‘Mr Fruity’ that the firm should
produce to maximise profits. You may assume that non-alcoholic cocktails are so
popular that the firm can sell all that it produces.
3. In a student’s diet a meal consists of beefburgers and chips. Beefburgers have 1 unit of
nutrient N1, 4 units of N2 and 125 calories per ounce. The figures for chips are 1/2 unit
of N1, 1 unit of N2 and 60 calories per ounce. In the interests of the student’s health it
is essential for the meal to contain at least 7 units of N1 and 22 units of N2.
Section 8.2 Applications of linear programming 613
What should the student ask for on the next visit to the refectory to satisfy the nutrient
requirements and minimise the number of calories?
4. An Italian restaurant offers a choice of pasta or pizza meals. It costs $3 to make a pasta
dish, which it sells for $13. The corresponding figures for pizzas are $2 and $10, respectively. The maximum number of meals that can be cooked in a week is 1200 and the
restaurant has a weekly cost budget of $3000. How many pasta and pizza dishes should
be cooked each week to maximise profit?
5. A small manufacturer produces two kinds of good, A and B, for which demand exceeds
capacity. The production costs for each item of A and B are $6 and $3, respectively, and
the corresponding selling prices are $7 and $4. In addition, transport costs are 20 cents
and 30 cents for each good of type A and B, respectively. The conditions of a bank loan
limit the manufacturer to maximum weekly production costs of $2700 and maximum
weekly transport costs of $120. How should the manufacturer arrange production to
maximise profit?
6. A publisher decides to use one section of its plant to produce two textbooks called
Microeconomics and Macroeconomics. The profit made on each copy is $12 for
Microeconomics and $18 for Macroeconomics. Each copy of Microeconomics requires
12 minutes for printing and 18 minutes for binding. The corresponding figures for
Macroeconomics are 15 and 9 minutes, respectively. There are 10 hours available for
printing and 10.5 hours available for binding. How many of each should be produced
to maximise profit?
7. A food producer uses two processing plants, P1 and P2, that operate seven days a week.
After processing, beef is graded into high-, medium- and low-quality foodstuffs. Highquality beef is sold to butchers, medium-quality beef is used in supermarket readymeals and the low-quality beef is used in dog food. The producer has contracted to
provide 120 kg of high-, 80 kg of medium- and 240 kg of low-quality beef each week.
It costs $4000 per day to run plant P1 and $3200 per day to run plant P2. Each day
P1 processes 60 kg of high-quality beef, 20 kg of medium-quality beef and 40 kg of
low-quality beef. The corresponding quantities for P2 are 20 kg, 20 kg and 120 kg,
respectively. How many days each week should the plants be operated to fulfil the beef
contract most economically?
8. [In this question, you are asked to formulate a linear programming problem. You are
NOT expected to solve it.]
An American university has enough places for 9000 students. Government restrictions
mean that at least 75% of the places must be given to US students but the remainder
may be given to non-US citizens. There are 5000 residential places available on campus.
All overseas students and at least one-quarter of the US students must be given
places on campus. The university gets $12 000 in tuition fees for each US student and
$15 000 for each overseas student. It wants to maximise the fees received.
Using the letter x for the number of places given to US students and y for the number
of places for overseas students,
(a) write down an expression for the objective function and state whether it is to be
maximised or minimised;
(b) write down the five constraints that define the feasible region and explain your
reasoning carefully;
(c) identify which aspect of the original problem has been overlooked in parts (a) and (b).
614 Chapter 8 Linear Programming
Exercise 8.2*
1. Leo has $12.50 to spend on his weekly supply of sweets, crisps and apples. A bag of
crisps costs $0.65, a bag of sweets costs $0.85, and one apple costs $0.50. The total
number of packets of crisps, sweets and apples consumed in a week must be at least
seven, and he eats at least twice as many packets of sweets as crisps. His new healthy
diet also means that total number of packets of sweets and crisps must not exceed onethird of the number of apples. If s, c and a denote the number of packets of sweets,
packets of crisps, and apples, respectively, which two of the following represent the
constraints defining the feasible region?
(1) 3c + 3s ≤ a
(2) s ≥ 0.5c
(3) 0.65s + 0.85c + 0.5a ≥ 12.5
(4) a + c + s > 7
(5) s ≤ c − a
(6) 17s + 10a + 13c ≤ 250
(7) c ≤ 2s
(8) a + c + s ≤ 7
(9) 3c + 3s + a ≤ 0
(10) 3s − 3c + a ≥ 0
2. A private hospital specialises in cosmetic and orthopaedic surgery. The profit made on
each patient undergoing cosmetic surgery is $12 000, and for orthopaedic surgery the
profit is $14 000. Each patient undergoing cosmetic surgery needs 1.5 hours of surgery
followed by 12 hours of post-operative care. Orthopaedic surgery takes 2 hours with
10 hours of post-operative care. Each week the hospital can provide a total of 18 hours
of surgical time and 120 hours of post-operative care. How many cosmetic and orthopaedic patients should the hospital accept each week to maximise total profit?
3. A firm manufactures two products, X and Y. To make 1 unit of product X
requires 3 units of raw materials and 2 units of labour. To make 1 unit of product Y
requires 5 units of raw materials and 2 units of labour. The total number of units
available for raw materials and labour is 31 500 and 17 000, respectively. The firm
makes a profit of $15 for making and selling product X. The corresponding profit for
Y is $20.
(a) Formulate a linear programming problem to maximise the firm’s profits.
(b) Solve the linear programming problem graphically.
(c) Find the shadow price of raw materials and give an interpretation of its value.
4. A manufacturer of outdoor clothing makes wax jackets and trousers. Each jacket
requires one hour to make, whereas each pair of trousers takes 40 minutes. The materials
for a jacket cost $32 and those for a pair of trousers cost $40. The company can devote
only 34 hours per week to the production of jackets and trousers, and the firm’s total
weekly cost for materials must not exceed $1200. The company sells the jackets at a
profit of $12 each and the trousers at a profit of $14 per pair. Market research indicates
Section 8.2 Applications of linear programming 615
that the firm can sell all of the jackets that are produced, but that it can sell at most half
as many pairs of trousers as jackets.
(a) How many jackets and trousers should the firm produce each week to maximise
profit?
(b) Due to the changes in demand, the company has to change its profit margin on a
pair of trousers. Assuming that the profit margin on a jacket remains at $12 and the
manufacturing constraints are unchanged, find the minimum and maximum profit
margins on a pair of trousers which the company can allow before it should change
its strategy for optimum output.
5. A farmer wishes to feed pigs with minimum cost but needs to ensure that each receives
at least 1.6 kg of protein, at least 0.3 kg of amino acid and no more than 0.3 kg of
calcium per day. Foods available are fish meal and meat scraps, which contain protein,
calcium and amino acid according to the following table:
Kg protein
per kg feed
Kg calcium
per kg feed
Kg amino acid
per kg feed
0.60
0.50
0.05
0.11
0.18
0.05
Fish meal
Meat scraps
Fish meal costs $0.65 per kg, whereas meat scraps cost $0.52 per kg. Determine a
minimum-cost feeding programme.
6. A individual has $300 000 to invest across a range of companies, A, B and C. The table
shows the expected percentage risk, dividend and growth in each company.
Company
A
B
C
% risk
% dividend
% growth
10
15
5
15
5
5
10
25
15
The individual wants to invest all of the $300 000 with minimum risk, but wants to
make sure that the expected dividend is at least 10% and growth is at least 15% of the
investment.
(a) If x and y denote the amount invested in company A and company B, respectively,
show that the problem may be formulated as:
Minimise
0.05x + 0.1y + 15 000
subject to
0.1x ≥ 15 000
−0.05x + 0.1y ≥ 0
x + y ≤ 300 000
x ≥ 0, y ≥ 0
(b) Solve the linear programming problem in part (a) to work out the optimal investment
in companies A, B and C.
7. A supplier of computer games has two warehouses, A and B, which are located in
different parts of the country. Warehouse A has 90 copies of a new computer game
in stock, whereas B has 60 copies. Customers in town C request 50 copies and those in
town D request 80 copies. The costs of delivering a copy to customers in towns C and
D are shown in the table.
➜
616 Chapter 8 Linear Programming
From/To
A
B
C
D
$1
$0.90
$1.30
$1.10
(a) If the supplier wishes to arrange delivery to minimise costs, show that the problem
can be written as a linear programming problem involving two variables.
(b) Draw the feasible region for the problem in part (a) and hence determine the optimal
delivery arrangements.
8. A company makes and sells two products, X and Y. The selling prices of each product
are $120 and $258, respectively. The resource requirements for making 1 item of each
product are listed in the table:
Product
Raw material 1
Raw material 2
Machine time
Labour time
X
Y
5 kg
1 kg
3 hours
3 hours
8 kg
1 kg
8 hours
10 hours
The unit cost of each resource and the maximum resource availability is shown in the
table:
Resource
Cost
Total available
Raw material 1
Raw material 2
Machine time
Labour time
$3 per kg
$5 per kg
$8 per hour
$12 per hour
6150 kg
1086 kg
5190 hours
5250 hours
(a) Formulate and solve a linear programming problem which will maximise profit.
(b) State the shadow price of machine time. Would you advise the company to increase
this resource?
Formal mathematics
We conclude this chapter by stating the general linear programming problem for n variables.
The problem involves maximising or minimising the objective function:
z = c1x1 + c2 x2 + . . . + cn xn
subject to m linear constraints:
a11x1 + a12 x2 + . . . + a1n xn ≤ b1
a21x1 + a22 x2 + . . . + a2n xn ≤ b2
...
am1x1 + am2 x2 + . . . + amn xn ≤ bm
and non-negativity constraints:
x1 ≥ 0, x2 ≥ 0, . . . . . . , xn ≥ 0
where ci, aij and bj are given constants.
If any of the linear constraints in a given problem are presented with a ‘≥’ sign, they can
always be re-cast in the above form by simply multiplying both sides by the number −1.
Two other formulations are possible. In sigma notation the problem is to maximise or
minimise
n
z 5 Sci xi subject to
i51
n
S
ai j xj # bi
j51
with
xj $ 0
and in matrix notation we maximise or minimise
cTx subject to Ax ≤ b with x ≥ 0
Multiple choice questions
Question 1
Show the set of points in the (x, y) plane which satisfy the inequality 3x − 4y ≥ 12. Use the
convention of shading the unwanted region.
y
y
4
4
x
x
23
23
A.
B.
y
y
3
x
24
x
24
23
C.
D.
y
x
23
24
E.
Multiple choice questions 619
Question 2
The diagram shows the feasible region defined by
3x + 2y ≥ 22
5x + 9y ≥ 55
6x + 2y ≥ 27
x≥0
y≥0
y
14
13
12
11
10
9
8
7
6
5
4
P
3
2
1
1
2
3
4
5
6
7
8
9
10
11
Find the exact coordinates of the point labelled P on the diagram.
A.
88 55
,
17 17
B.
26 16
,
5 5
C.
53 61
,
10 20
D.
68 125
,
13 39
E.
646 81
,
125 25
x
620 Chapter 8 Linear Programming
Question 3
Find the number of points with integer coordinates which lie in the feasible region defined
by:
3x + 8y ≤ 24
x+y≤5
x≥1
y≥0
A. 8
B. 9
C. 10
D. 11
E. 12
Questions 4–6 all relate to the following diagram which shows the feasible region defined
by the following inequalities:
x + 2y ≤ 17
5x + 3y ≤ 36
x≤6
y≤8
x≥0
y≥0
y
8
7
6
5
4
3
2
1
1
2
3
4
5
6
7
x
Question 4
Find the maximum value of the objective function x + y in the feasible region.
A. 8
B. 9
C. 10
D. 11
E. 12
Multiple choice questions 621
Question 5
Find the minimum value of the objective function, −x − 1.5y in the feasible region.
A. −13
B. −15.5
C. −13.5
D. −12
E. −14
Question 6
The linear programming problem is to maximise 2x + ay in the feasible region and the
solution set consists of all points on the line segment joining (1, 8) and (3, 7). Find the value
of a.
A. 0
B. 1
C. 2
D. 3
E. 4
Questions 7–10 relate to the following linear programming problem:
A dietician recommends that a particular individual must consume a minimum of 18 units
of calcium, 16 units of iron and 14 units of zinc each week. The person would like to make
sure that she complies with the diet by buying some food supplements containing all the
nutrients she needs from her local health shop which sells packets of ‘VitaPlus’ and
‘BeHealthy’. She would like to choose a viable combination of these supplements at minimal cost. VitaPlus costs $3 a packet and contains 1 unit of calcium, 4 units of iron and 1
unit of zinc. A packet of BeHealthy costs $4 and contains 1.5 units of calcium, 1 unit of
iron and 1 unit of zinc.
Letters x and y denote the number of packets of VitaPlus and BeHealthy, respectively.
Question 7
Write down the objective function and state whether it is to be maximised or minimised.
A. Minimise 4x + 3y
B. Maximise 3x + 4y
C. Minimise 3x − 4y
D. Minimise 3x + 4y
E. Maximise 4x + 3y
622 Chapter 8 Linear Programming
Question 8
Write down the inequalities defining the constraints
A. 2x + 3y ≥ 36, 4x + y ≥ 16, x + y ≥ 14, x ≥ 0, y ≥ 0
B. 2x + 3y ≤ 36, 4x + y ≤ 16, x + y ≤ 14, x ≥ 0, y ≥ 0
C. 3x + 2y ≥ 36, x + 4y ≥ 16, x + y ≥ 14, x ≥ 0, y ≥ 0
D. 3x + 2y ≤ 36, 4x + y ≤ 16, x + y ≤ 14, x ≥ 0, y ≥ 0
E. 2x + 3y ≤ 36, x + 4y ≥ 16, x + y ≥ 14, x ≥ 0, y ≥ 0
Question 9
Use the feasible region sketched in the diagram below to work out the number of packets
of VitaPlus and BeHealthy that need to be bought.
A. 18 packets of VitaPlus and 0 packets of BeHealthy
B. 1 packet of VitaPlus and 13 packets of BeHealthy
C. 1 packet of VitaPlus and 14 packets of BeHealthy
D. 0 packets of VitaPlus and 16 packets of BeHealthy
E. 6 packets of VitaPlus and 8 packets of BeHealthy
y
18
16
14
12
10
8
6
4
2
2
4
Question 10
Calculate the shadow price of iron.
A. 1/3
B. 5/3
C. 3/5
D. 0
E. 3
6
8
10
12
14
16
18
x
Examination questions
Question 1
(a) Solve the linear programming problem:
Maximise
x+y
subject to
x + 4y ≤ 12
2x + y ≤ 10
x≥0
y≥0
(b) Find the solution if the objective is to
(i) minimise x − 3y
1
(ii) maximise x 1 y
2
subject to the same constraints as before.
Question 2
Solve the linear programming problem:
Minimise
6x + 5y
subject to
x+y≥8
x + 2y ≥ 10
3x + y ≥ 12
x≥0
y≥0
What can you say about the solution if the objective is to maximise 5x + 2y?
Question 3
A designer clothing firm makes ladies’ scarves and blouses using a blend of wool and silk
fibres. The quantity of fibre (in grams) required for a scarf and a blouse, and the profit
(in $s) made from the sale of each item are shown in the table below:
Silk
Wool
Profit
Scarf
12
24
70
Blouse
18
12
84
624 Chapter 8 Linear Programming
(a) Write down an expression for the total profit if the firm makes x scarves and y blouses.
(b) If the firm has a weekly supply of 10.8 kg of silk and 19.2 kg of wool available, write
down four constraints on the variables x and y.
(c) Show the feasible region on a sketch graph.
(d) How many scarves and blouses should the firm make each week to maximise profit?
Question 4
A specialist bakery makes two types of cakes, chocolate and fruit. Each chocolate cake
uses 600 grams of flour, 100 grams of butter and 200 grams of sugar. The corresponding
figures for a fruit cake are 300 grams, 100 grams and 300 grams. The bakery makes a profit
of $20 and $24 for each chocolate and fruit cake sold. The bakery wants to make as many
cakes of each type that it can to maximise total profit. However, it only has 96 kg flour,
20 kg butter and 52 kg sugar available each week. You may assume that there are no other
restrictions and that the bakery can sell all cakes that it makes.
(a) Give a precise description of the two decision variables, x and y.
(b) Write down an expression in terms of x and y for the weekly profit.
(c) Write down inequalities for all constraints in the problem.
(d) Draw the feasible region and identify the coordinates of all of its corners.
(e) How many cakes of each type should be baked each week to maximise profit?
Question 5
I decide to extend my investment portfolio by spending no more than $5000 on shares in
companies A and B. The cost of each share is $5 and $10, respectively, and my financial
advisor estimates that I can expect to make a profit of 80 cents and 65 cents on each share.
Due to environmental concerns, I decide to buy at least twice as many shares in company
B as in company A.
(a) Formulate this as a linear programming problem.
(b) Sketch the feasible region.
(c) Write down the coordinates of the corners of the feasible region.
(d) How many of each type of share should I buy? State the corresponding profit.
Question 6
A fruit farmer supplies raspberries and blueberries to a leading supermarket. She is contracted to supply at least 100 kg of raspberries and 50 kg blueberries each week. The supermarket requires that the total weight of fruit delivered must be at least 200 kg. In a typical
week she can harvest a maximum of 200 kg of raspberries and 300 kg of blueberries.
The cost of supplying a kilo of raspberries is $4.40, and the corresponding figure for
blueberries is $3.10.
(a) Formulate this as a linear programming problem if the farmer’s strategy is to minimise
total cost.
Examination questions 625
(b) Sketch the feasible region.
(c) Write down the coordinates of the corners of the feasible region.
(d) How many kilos of raspberries and blueberries should the farmer supply each week to
the supermarket?
(e) The selling prices of a kilo of raspberries and a kilo of blueberries are $8 and $10,
respectively. Work out the maximum weekly profit that the supermarket could make.
Question 7
(a) Find the inverse of the matrix
4 3
.
5 7
Hence find the exact solution of the simultaneous equations
4x + 3y = 900
5x + 7y = 1750
(b) A food supplier makes two different varieties of smoothie, ‘Exotic’ and ‘Floridian’,
which it sells in 2-litre cartons. The main ingredients and the profit made on each carton are as follows:
Banana
Mango
Orange
Profit
Exotic
4
2
5
$4.50
Floridian
3
1
7
$4.00
The company has 900 bananas, 400 mangoes and 1750 oranges available and wishes
to arrange production to maximise total profit. It can sell all of the smoothies that it
produces.
(i) Formulate this as a linear programming problem.
(ii) Sketch the feasible region.
(iii) Write down the exact coordinates of the corners of the feasible region.
(iv) Find the optimising production levels of each type of smoothie.
626  
Chapter 9
Dynamics
This chapter provides a simple introduction to the dynamics of economic systems. So far, all of
our models have been static. We have implicitly assumed that equilibrium prices and incomes are
somehow attained instantaneously, as if by magic. In practice, economic variables take time to
vary. The incorporation of time into economic models is known as dynamics, and it enables us
to decide whether equilibrium values are actually achieved and, if so, exactly how individual
variables approach these values. There are two sections that can be read in either order.
In Section 9.1 we consider the case when time, t, is a discrete variable taking whole number
values. This describes the situation in which variables change only after fixed periods. For
example, the prices of certain agricultural products change from one season to the next but are
fixed throughout each season. We express this time dependence using an obvious subscript
notation. The price during the first period is denoted by P1, the price during the second period
is P2 and so on. Equations that relate the price in one period, Pt, to that of the previous period,
Pt−1, are called difference equations, and a method is described for the solution of such
equations.
In Section 9.2 we consider the case when time is a continuous variable taking all possible values
in a certain interval. This describes the situation in which variables change from one moment to
the next. For example, the prices of certain commodities, such as oil, effectively change instantane­
ously and are not fixed on a seasonal basis. We express this time dependence using the usual
function notation, P(t). It is appropriate to model the rate of change of P with respect to t using
the derivative, P′(t). Equations that involve the derivatives of an unknown function are called
differential equations, and a method is described for solving such equations.
It is not possible in an introductory text such as this to give you more than a flavour of the
mathematics of dynamics. However, in spite of this, we show you how to solve dynamic systems
in both macroeconomics and microeconomics. Also, we hope that this chapter will encourage
you to read other texts that describe more advanced methods and models.
Section 9.1
Difference equations
Objectives
At the end of this section you should be able to:
●
Find the complementary function of a difference equation.
●
Find the particular solution of a difference equation.
●
Analyse the stability of economic systems.
●
Solve lagged national income determination models.
●
Solve single-commodity-market models with lagged supply.
A difference equation (sometimes called a recurrence relation) is an equation that relates
consecutive terms of a sequence of numbers. For example, the equation
Yt = 2Yt−1
describes sequences in which one number is twice its predecessor. There are obviously many
sequences that satisfy this requirement, including
2, 4, 8, 16, . . .
5, 10, 20, 40, . . .
−1, −2, −4, −8, . . .
In order to determine the sequence uniquely, we need to be given some additional information, such as the first term. It is conventional to write the first term as Y0, and once this is
given a specific value, all remaining terms are known.
If Y0 = 3, then
Y1 = 2Y0 = 2 × 3 = 6
Y2 = 2Y1 = 2 × 6 = 12
Y3 = 2Y2 = 2 × 12 = 24
If we write these terms as
Y1 = 2Y0 = 21 × 3
Y2 = 2Y1 = 22 × 3
Y3 = 2Y2 = 23 × 3
we can see that the general term may be expressed as
Yt = 3(2t )
Section 9.1 Difference equations 629
This is the solution of the difference equation
Yt = 2Yt−1
with initial condition Y0 = 3.
The following problem gives you an opportunity to solve difference equations for
yourself.
Practice Problem
1. Starting with the given initial conditions, write down the first four terms of each of the
following sequences. By expressing these as an appropriate power, write down a
formula for the general term Yt in terms of t.
(1) (a) Yt = 3Yt−1; Y0 = 1
(2) (a) Yt 5
1
Y ;Y 51
2 t 21 0
(b) Yt = 3Yt−1; Y0 = 7
(b) Yt 5
1
Y ;Y 57
2 t 21 0
(c) Yt = 3Yt−1; Y0 = A
(c) Yt 5
1
Y ;Y 5A
2 t 21 0
(3) Yt = bYt−1; Y0 = A
The result of the last part of Practice Problem 1 shows that the solution of the general
equation
Yt = bYt−1
(1)
with initial condition
Y0 = A
is given by
Yt = A(bt )
Before we can consider the use of difference equations in economic models, we must
examine the solution of more general equations of the form
Yt = bYt−1 + c
(2)
where the right-hand side now includes a non-zero constant, c. We begin by defining some
terminology. The general solution of equation (2) can be written as the sum of two separate
expressions known as the complementary function (CF) and the particular solution (PS).
The complementary function is the name that we give to the solution of equation (2) when
the constant, c, is zero. In this case, equation (2) reduces to equation (1), and so
CF = A(bt )
The particular solution is the name that we give to any solution of equation (2) that we are
clever enough to ‘spot’. This turns out to be easier to do than might at first appear, and we
will see how this can be done in a moment. Finally, once CF and PS have been found, we can
write down the general solution of equation (2) as
Yt = CF + PS = A(bt ) + PS
The letter A is no longer equal to the first term, Y0, although it can easily be calculated, as
the following example demonstrates.
630 Chapter 9 Dynamics
Example
Solve the following difference equations with the specified initial conditions. Comment on
the qualitative behaviour of the solution in each case.
(a) Yt = 4Yt−1 + 21; Y0 = 1
(b) Yt 5
1
Y 1 8; Y0 5 2
3 t 21
Solution
(a) The difference equation
Yt = 4Yt−1 + 21
is of the standard form
Yt = bYt−1 + c
and so can be solved using the complementary function and particular solution. The
complementary function is the general solution of the equation when the constant term
on the right-hand side is replaced by zero: that is, it is the solution of
Yt = 4Yt−1
which is A(4t ).
The particular solution is any solution of the original equation
Yt = 4Yt−1 + 21
that we are able to find. In effect, we need to think of a sequence of numbers, Yt , such
that when this is substituted into
Yt − 4Yt−1
we obtain the constant value of 21. One obvious sequence likely to work is a constant
sequence,
Yt = D
for some number, D. If this is substituted into
Yt = 4Yt−1 + 21
we obtain
D = 4D + 21
(Note that Yt = D whatever the value of t, so Yt−1 is also equal to D.) This algebraic
equation can be rearranged to get
−3D = 21
and so D = −7.
We have therefore shown that the complementary function is given by
CF = A(4t )
and that the particular solution is
PS = −7
Section 9.1 Difference equations 631
Hence
Yt = CF + PS = A(4t ) − 7
which is the general solution of the difference equation
Yt = 4Yt−1 + 21
To find the specific solution that satisfies the initial condition
Y0 = 1
we simply put t = 0 in the general solution to get
Y0 = A(40 ) − 7 = 1
that is,
A−7=1
which gives
A=8
The solution is
Yt = 8(4t ) − 7
A graphical interpretation of this solution is shown in Figure 9.1, where Yt is plotted
against t. It is tempting to join the points up with a smooth curve. However, this does
Yt
2000
1800
1600
1400
Yt 5 8(4 t ) 2 7
1200
1000
800
600
400
200
0
Figure 9.1
1
2
3
4
5
t
➜
632 Chapter 9 Dynamics
not make sense because t is allowed to take only whole number values. Consequently,
we join up the points with horizontal lines to create the ‘staircase’ which more properly
reflects the fact that t is discrete. Figure 9.1 shows that the values of Yt increase
without bound as t increases. This is also apparent from the formula for Yt because the
numbers 4t get ever larger as t increases. We describe this by saying that the time path
diverges uniformly or explodes. This sort of behaviour can be expected to occur for any
solution
Yt = A(bt ) + PS
where b > 1.
(b) The difference equation
Yt 5
1
Y 18
3 t 21
can be solved in a similar way to that of part (a). The complementary function is
given by
CF 5 A
1
3
t
and for a particular solution we try
Yt = D
for some constant D. Substituting this into the difference equation gives
D5
1
D18
3
which has solution D = 12, so
PS = 12
The general solution is therefore
t
Yt 5 CF 1 PS 5 A
1
1 12
3
Finally, the specific value of A can be found from the initial condition
Y0 = 2
Setting t = 0 in the general solution gives
25A
1
3
0
1 12 5 A 1 12
and so A is −10. The solution is
t
1
Yt 5 210
1 12
3
This solution is sketched in Figure 9.2, which shows that the values of Yt increase
but eventually settle down at 12. We describe this by saying that the time path
converges uniformly to the value of 12, which is referred to as the equilibrium value. This
Section 9.1 Difference equations 633
Yt
equilibrium value
12
11
10
9
8
7
6
1
Yt 5 210( ) t 1 12
3
5
4
3
2
1
0
1
2
3
4
5
t
Figure 9.2
behaviour is also apparent from the formula for Yt because the number (1/3)t gets ever
smaller as t increases. In effect, the complementary function disappears, leaving just the
particular solution. The particular solution is the equilibrium value of Yt, whereas the
complementary function measures the deviation from the equilibrium which, in this
case, converges to zero as t increases. This sort of behaviour can be expected to occur
for any solution
Yt = A(bt ) + PS
when 0 < b < 1.
Practice Problem
2. Solve the following difference equations with the specified initial conditions:
(a) Yt = −1/2Yt−1 + 6; Y0 = 0
(b) Yt = −2Yt−1 + 9; Y0 = 4
In each case, sketch the corresponding ‘staircase’ diagram and comment on the qualitative
behaviour of the solution as t increases.
634 Chapter 9 Dynamics
The results of the previous example and Practice Problem 2 can be summarised:
If b > 1, then Yt displays uniform divergence.
If 0 < b < 1, then Yt displays uniform convergence.
If −1 < b < 0, then Yt displays oscillatory convergence.
If b < −1, then Yt displays oscillatory divergence.
The remaining possibilities, b = 1, b = −1 and b = 0, are considered in Problem 1 of Exercise 9.1
at the end of this section, which shows that Yt converges when b is 0 but diverges when b is
1 or −1. We conclude that the solution of the difference equation eventually settles down to
an equilibrium state only when b lies in the range −1 < b < 1.
If convergence does occur in an economic model, the model is said to be stable. If the
variables diverge, it is said to be unstable.
We now investigate two applications of difference equations taken from macroeconomics
and microeconomics, respectively:
●
national income determination;
●
supply and demand analysis.
We consider each of these in turn.
9.1.1 National income determination
In Section 1.7 we introduced a simple two-sector model with structural equations
Y=C+I
C = aY + b
I = I*
where b and I* denote autonomous consumption and investment, and a is the marginal propensity to consume, which lies in the range 0 < a < 1. In writing down the equations in this
form, we are implicitly assuming that only one time period is involved, that consumption
depends on national income within this time period and that equilibrium values are attained
instantaneously. In practice, there is a time lag between consumption and national income.
Consumption, Ct, in period t depends on national income, Yt−1, in the previous period, t − 1.
The corresponding consumption function is given by
Ct = aYt−1 + b
If we assume that investment is the same in all time periods, then
It = I*
for some constant, I*. Finally, if the flow of money is in balance in each time period, we also
have
Yt = Ct + It
Substituting the expressions for Ct and It into this gives
Yt = aYt−1 + b + I*
which we recognise as a difference equation of the standard form given in this section. This
equation can therefore be solved and the time path analysed.
Section 9.1 Difference equations 635
Example
Consider a two-sector model:
Yt = Ct + It
Ct = 0.8Yt−1 + 100
It = 200
Find an expression for Yt when Y0 = 1700. Is this system stable or unstable?
Solution
Substituting the expressions for Ct and It into
Yt = Ct + It
gives
Yt = (0.8Yt−1 + 100) + 200
= 0.8Yt−1 + 300
The complementary function is given by
CF = A(0.8)t
and for a particular solution we try
Yt = D
for some constant D. Substituting this into the difference equation gives
D = 0.8D + 300
which has solution D = 1500. The general solution is therefore
Yt = A(0.8)t + 1500
The initial condition,
Y0 = 1700
gives
1700 = A(0.8)0 + 1500 = A + 1500
and so A is 200. The solution is
Yt = 200(0.8t) + 1500
As t increases, (0.8)t converges to zero, and so Yt eventually settles down at the
equilibrium level of 1500. The system is therefore stable. Note also that because 0.8 lies
between 0 and 1, the time path displays uniform convergence.
Practice Problem
3. Consider the two-sector model:
Yt = Ct + It
Ct = 0.9Yt−1 + 250
It = 350
Find an expression for Yt when Y0 = 6500. Is this system stable or unstable?
636 Chapter 9 Dynamics
In the previous example and again in Practice Problem 3, we noted that the model is stable
and that it displays uniform convergence. If we return to the general equation
Yt = aYt−1 + b + I*
it is easy to see that this is always the case for the simple two-sector model because the
coefficient of Yt−1 is the marginal propensity to consume, which is known to lie between 0 and 1.
9.1.2 Supply and demand analysis
In Section 1.5 we introduced a simple model of supply and demand for a single good in an
isolated market. If we assume that the supply and demand functions are both linear, then we
have the relations
QS = aP − b
QD = −cP + d
for some positive constants, a, b, c and d. (Previously, we have written P in terms of Q and
have sketched the supply and demand curves with Q on the horizontal axis and P on the
vertical axis. It turns out that it is more convenient in the present context to work the other
way round and to write Q as a function of P.) In writing down these equations, we are
implicitly assuming that only one time period is involved, that supply and demand are
dependent only on the price in this time period and that equilibrium values are attained
instantaneously. However, for certain goods, there is a time lag between supply and price.
For example, a farmer needs to decide precisely how much of any crop to sow well in advance
of the time of sale. This decision is made on the basis of the price at the time of planting and
not on the price prevailing at harvest time, which is unknown. In other words, the supply,
QSt, in period t depends on the price, Pt−1, in the preceding period t − 1. The corresponding
time-dependent supply and demand equations are
QSt = aPt−1 − b
QDt = −cPt + d
If we assume that, within each time period, demand and supply are equal, so that all goods
are sold, then
QDt = QSt
that is,
−cPt + d = aPt−1 − b
This equation can be rearranged as
−cPt = aPt−1 − b − d
Pt 5 2
a
b1d
P 1
c t21
c
(subtract d from both sides)
(divide both sides by −c)
which is a difference equation of the standard form. The equation can therefore be solved in
the usual way and the time path analysed. Once a formula for Pt is obtained, we can use the
demand equation
Qt = −cPt + d
to deduce a corresponding formula for Qt by substituting the expression for Pt into the
right-hand side.
Section 9.1 Difference equations 637
Example
Consider the supply and demand equations
QSt = 4Pt−1 − 10
QDt = −5Pt + 35
Assuming that the market is in equilibrium, find expressions for Pt and Qt when P0 = 6.
Is the system stable or unstable?
Solution
If
QDt = QSt
then
−5Pt + 35 = 4Pt−1 − 10
which rearranges to give
−5Pt = 4Pt−1 − 45
Pt = −0.8Pt−1 + 9
(subtract 35 from both sides)
(divide both sides by −5)
The complementary function is given by
CF = A(−0.8)t
and for a particular solution we try
Pt = D
for some constant D. Substituting this into the difference equation gives
D = −0.8D + 9
which has solution D = 5. The general solution is therefore
Pt = A(−0.8)t + 5
The initial condition, P0 = 6, gives
6 = A(−0.8)0 + 5 = A + 5
and so A is 1. The solution is
Pt = (−0.8)t + 5
An expression for Qt can be found by substituting this into the demand equation
Qt = −5Pt + 35
to get
Qt = −5[(−0.8)t + 5] + 35
= −5(−0.8)t + 10
As t increases, (− 0.8)t converges to zero, and so Pt and Qt eventually settle down at the
equilibrium levels of 5 and 10, respectively. The system is therefore stable. Note also that
because − 0.8 lies between −1 and 0, the time paths display oscillatory convergence.
638 Chapter 9 Dynamics
Practice Problems
4. Consider the supply and demand equations
QSt = Pt−1 − 8
QDt = −2Pt + 22
Assuming equilibrium, find expressions for Pt and Qt when P0 = 11. Is the system stable
or unstable?
5. Consider the supply and demand equations
QSt = 3Pt−1 − 20
QDt = −2Pt + 80
Assuming equilibrium, find expressions for Pt and Qt when P0 = 8. Is the system stable
or unstable?
Two features emerge from the previous example and Practice Problems 4 and 5. Firstly,
the time paths are always oscillatory. Secondly, the system is not necessarily stable, and
so equilibrium might not be attained. These properties can be explained if we return to the
general equation
Pt 5 2
a
b1d
Pt21 1
c
c
The coefficient of Pt−1 is −a/c. Given that a and c are both positive, it follows that −a/c is
negative and so oscillations will always be present. Moreover,
●
if a > c, then −a/c < −1 and Pt diverges
●
if a < c, then −1 < −a/c < 0 and Pt converges.
We conclude that stability depends on the relative sizes of a and c, which govern the slopes
of the supply and demand curves. Bearing in mind that we have chosen to consider supply
and demand equations in which Q is expressed in terms of P, namely
QS = aP − b
QD = −cP + d
we deduce that the system is stable whenever the supply curve is flatter than the demand
curve when P is plotted on the horizontal axis.
Throughout this section we have concentrated on linear models. An obvious question to
ask is whether we can extend these to cover the case of non-linear relationships. Unfortunately,
the associated mathematics gets hard very quickly, even for mildly non-linear problems. It is
usually impossible to find an explicit formula for the solution of such difference equations.
Under these circumstances, we fall back on the tried and trusted approach of actually calculating the first few values until we can identify its behaviour. A spreadsheet provides an ideal
way of doing this, since the parameters in the model can be easily changed.
Section 9.1 Difference equations 639
Key Terms
Complementary function of a difference equation The solution of the difference equation
Yt = bYt−1 + c when the constant c is replaced by zero.
Difference equation An equation that relates consecutive terms of a sequence of numbers.
Dynamics
Analysis of how equilibrium values vary over time.
Equilibrium value of a difference equation A solution of a difference equation that does not vary
over time; it is the limiting value of Yn as n tends to infinity.
General solution of a difference equation The solution of a difference equation that con­tains an
arbitrary constant. It is the sum of the complementary function and a particular solution.
Initial condition The value of Y0 that needs to be specified to obtain a unique solution of a differ­
ence equation.
Particular solution of a difference equation
Yt = bYt−1 + c.
Any one solution of a difference equation such as
Recurrence relation An alternative term for a difference equation. It is an expression for Yn in
terms of Yn−1 (and possibly Yn−2, Yn−3, etc.).
Stable (unstable) equilibrium An economic model in which the solution of the associated differ­
ence equation converges (diverges).
Uniformly convergent sequence
decreases) to a finite limit.
A sequence of numbers which progressively increases (or
Uniformly divergent sequence A sequence of numbers which progressively increases (or
decreases) without a finite limit.
Exercise 9.1
1. Calculate the first four terms of the sequences defined by the following difference
equations. Hence write down a formula for Yt in terms of t. Comment on the qualitative behaviour of the solution in each case.
(a) Yt = Yt−1 + 2; Y0 = 0
(b) Yt = −Yt−1 + 6; Y0 = 4 (c) Yt = 0Yt−1 + 3; Y0 = 3
2. Solve the following difference equations with the specified initial conditions:
1
(a) Yt 5 Yt21 1 6; Y0 5 1
4
(b) Yt = −4Yt−1 + 5; Y0 = 2
Comment on the qualitative behaviour of the solution as t increases.
3. Consider the two-sector model:
Yt = Ct + It
Ct = 0.7Yt−1 + 400
It = 0.1Yt−1 + 100
Given that Y0 = 3000, find an expression for Yt . Is this system stable or unstable?
4. Consider the supply and demand equations
QSt = 0.4Pt−1 − 12
QDt = −0.8Pt + 60
➜
640 Chapter 9 Dynamics
Assuming that equilibrium conditions prevail, find an expression for Pt when P0 = 70.
Is the system stable or unstable?
5. Consider the two-sector model:
Yt = Ct + It
Ct = 0.75Yt−1 + 400
It = 200
Find the value of C2, given that, Y0 = 400.
6. The Harrod–Domar model of the growth of an economy is based on three assumptions.
(1) Savings, St, in any time period are proportional to income, Yt, in that period, so that
St = aYt
(a > 0)
(2) Investment, It, in any time period is proportional to the change in income from the
previous period to the current period so that
It = b(Yt − Yt−1) (b > 0)
(3) Investment and savings are equal in any period so that
It = St
Use these assumptions to show that
Y1 5
b
Yt21
b2a
and hence write down a formula for Yt in terms of Y0. Comment on the stability of
the system in the case when a = 0.1 and b = 1.4.
Exercise 9.1*
1. Describe the qualitative behaviour of the sequence of numbers which satisfy
1
Yt 5 2 Yt221
2
with initial condition Y0 = −1.
2. (a) Write down the next four terms of the sequence defined by
Yt 5
1
;Y 52
1 2 Yt21 0
Deduce the value of Y200.
(b) Write down the next four terms of the sequence defined by
Yt = Yt −1 + 4; Y0 = 3
Write down a formula for Yt in terms of t.
Section 9.1 Difference equations 641
3. Find the solution of the difference equation
Yt = bYt−1 + c
with initial condition, Y0 = a.
4. Consider the two-sector model:
Yt = Ct + It
Ct = 0.85Yt−1 + 300
It = 0.15Yt−1 + 100
Given that Y0 = 4000, find an expression for Yt. Is this system stable or unstable?
5. Consider the supply and demand equations
QS = aPt −1 − b
t
QD = − cPt + d
t
where the constants a, b, c and d are all positive.
(a) Assuming that the market is in equilibrium, show that
Pt 5 2
a
b1d
P 1
c t21
c
(b) Show that
b1d
is a particular solution of the difference equation in part (a), and
a1c
write down an expression for the general solution.
(c) State the conditions under which the solution in part (a) is guaranteed to converge,
and state the equilibrium price and quantity. Simplify your answers.
6. Find the reduced form connecting Pt and Pt−1 for the market model:
QSt = aPt − b
QDt = −cPt + d
Pt = Pt−1 − e(QSt−1 − QD t−1)
7. A bank offers an individual a loan of $A at a rate of interest of r% compounded
monthly. The individual pays off the loan by monthly instalments of $a. If ut denotes
outstanding balance after t months, explain why
ut 5 1 1
r
u 2a
1200 t21
and show that
ut 5 A 2
1200a
r
11
r
1200
t
1
1200a
r
Hence write down an expression for the monthly repayment if the loan is paid off after
N months.
➜
642 Chapter 9 Dynamics
8. Consider the difference equation
Yt = 0.1Yt−1 + 5(0.6)t
(a) Write down the complementary function.
(b) By substituting Yt = D(0.6)t into this equation, find a particular solution.
(c) Use your answers to parts (a) and (b) to write down the general solution and hence
find the specific solution that satisfies the initial condition, Y0 = 9.
(d) Is the solution in part (c) stable or unstable?
9. Consider the difference equation
Yt = 0.2Yt−1 + 0.8t + 5
(a) Write down the complementary function.
(b) By substituting Yt = Dt + E into this equation, find a particular solution.
(c) Use your answers to parts (a) and (b) to write down the general solution and hence
find the specific solution that satisfies the initial condition, Y0 = 10.
(d) Is the solution in part (c) stable or unstable?
Section 9.2
Differential equations
Objectives
At the end of this section you should be able to:
●
Find the complementary function of a differential equation.
●
Find the particular solution of a differential equation.
●
Analyse the stability of economic systems.
●
Solve continuous time national income determination models.
●
Solve continuous time supply and demand models.
A differential equation is an equation that involves the derivative of an unknown function.
Several examples have already been considered in Chapter 6. For instance, in Section 6.2 we
noted the relationship
dK
5I
dt
where K and I denote capital stock and net investment, respectively. Given any expression for
I(t), this represents a differential equation for the unknown function, K(t). In such a simple
case as this, we can solve the differential equation by integrating both sides with respect to t.
For example, if I(t) = t, the equation becomes
dK
5t
dt
and so
#
K(t) 5 tdt 5
t2
1c
2
where c is a constant of integration. The function K(t) is said to be the general solution of the
differential equation, and c is referred to as an arbitrary constant. Some additional information is needed if the solution is to be pinned down uniquely. This is usually provided in the
form of an initial condition in which we specify the value of K at t = 0. For example, the
capital stock may be known to be 500 initially. Substituting t = 0 into the general solution
K(t) 5
t2
1c
2
gives
K(0) 5
02
1 c 5 500
2
644 Chapter 9 Dynamics
and so c is 500. Hence the solution is
K(t) 5
t2
1 500
2
In this section we investigate more complicated equations where the right-hand sides of
these equations are given in terms of y such as
dy
5 3y
dt
The solution of this equation is any function, y(t), which differentiates to three times itself.
We have seen in Section 4.8 that emt differentiates to m times itself, so an obvious candidate
for the solution is
y = e3t
However, there are many functions with the same property, including
y = 2e3t, y = 5e3t
and
y = −7.52e3t
Indeed, any function of the form
y = Ae3t
satisfies this differential equation because
dy
5 3(Ae3t) 5 3y
dt
The precise value of the constant A is determined from an initial condition such as
y(0) = 5
If we substitute t = 0 into the general solution
y(t) = Ae3t
we get
y(0) = Ae0 = A
and so A is 5. The solution is
y(t) = 5e3t
Practice Problem
1. (a) Find the solution of the differential equation
dy
5 4y
dt
which satisfies the initial condition, y(0) = 6.
(b) Find the solution of the differential equation
dy
5 25y
dt
which satisfies the initial condition, y(0) = 2.
Section 9.2 Differential equations 645
Consider the differential equation
dy
5 my 1 c
dt
(1)
where m and c are constants. The general solution of equation (1) is the sum of two separate
functions, known as the complementary function (CF) and particular solution (PS). These
are defined in much the same way as their counterparts for difference equations discussed in
the previous section. The complementary function is the solution of equation (1) when the
constant term on the right-hand side is replaced by zero. In other words, the complementary
function is the solution of
dy
5 my
dt
The results of Practice Problem 1 show that this is given by
CF = Aemt
The particular solution is any solution that we are able to find of the original equation (1).
This can be done by ‘guesswork’, just as we did in Section 9.1. Finally, once CF and PS have
been determined, the general solution of equation (1) can be written down as
y = CF + PS = Aemt + PS
As usual, the specific value of A can be worked out at the very end of the calculations via
an initial condition.
Example
Solve the differential equation
dy
5 22y 1 100
dt
in the case when the initial condition is
(a) y(0) = 10
(b) y(0) = 90 (c) y(0) = 50
Comment on the qualitative behaviour of the solution in each case.
Solution
The differential equation
dy
5 22y 1 100
dt
is of the standard form
dy
5 my 1 c
dt
and so can be solved using the complementary function and particular solution.
➜
646 Chapter 9 Dynamics
The complementary function is the general solution of the equation when the constant
term is taken to be zero: that is, it is the solution of
dy
5 22y
dt
which is Ae−2t. The particular solution is any solution of the original equation
dy
5 22y 1 100
dt
that we are able to find. In effect, we need to think of a function, y(t), such that when it is
substituted into
dy
1 2y
dt
we obtain the constant value of 100. One obvious function likely to work is a constant
function,
y(t) = D
for some constant D. If this is substituted into
dy
5 22y 1 100
dt
we obtain
0 = −2D + 100
(Note that dy/dt = 0 because constants differentiate to zero.) This algebraic equation can
be rearranged to get
2D = 100
and so D = 50.
We have therefore shown that the complementary function is given by
CF = Ae−2t
and that the particular solution is
PS = 50
Hence
y(t) = CF + PS = Ae−2t + 50
This is the general solution of the differential equation
dy
5 22y 1 100
dt
(a) To find the specific solution that satisfies the initial condition
y(0) = 10
we simply put t = 0 into the general solution to get
y(0) = Ae0 + 50 = 10
Section 9.2 Differential equations 647
that is
A + 50 = 10
which gives
A = −40
The solution is
y(t) = −40e−2t + 50
A graph of y against t is sketched as the bottom graph in Figure 9.3. This shows
that y(t) increases from its initial value of 10 and settles down at the value of 50 for
sufficiently large t. As usual, this limit is called the equilibrium value and is equal to
the particular solution. The complementary function measures the deviation from the
equilibrium.
(b) If the initial condition is
y(0) = 90
then we can substitute t = 0 into the general solution
y(t) = Ae−2t + 50
to get
y(0) = Ae0 + 50 = 90
which has solution A = 40. Hence
y(t) = 40e−2t + 50
A graph of y against t is sketched as the top curve in Figure 9.3. In this case
y(t) decreases from its initial value of 90 but again settles down at the equilibrium level
of 50.
y
90
y(t) 5 40e22t 1 50
y(t) 5 50
50
y(t) 5 240e22t 1 50
10
t
Figure 9.3
➜
648 Chapter 9 Dynamics
(c) If the initial condition is
y(0) = 50
then we can substitute t = 0 into the general solution
y(t) = Ae−2t + 50
to get
y(0) = Ae0 + 50 = 50
which has solution A = 0. Hence
y(t) = 50
A graph of y against t is sketched as the horizontal line in Figure 9.3. In this case, y
is initially equal to the equilibrium value and y remains at this constant value for all
time.
Notice that the solution y(t) eventually settles down at the equilibrium value irrespective of the initial conditions. This is because the coefficient of t in the expression
CF = Ae−2t
is negative, causing CF to converge to zero as t increases. We would expect convergence to occur for any solution
y(t) = Aemt + D
when m < 0.
Practice Problem
2. Solve the following differential equation subject to the given initial condition.
Comment on the qualitative behaviour of the solution as t increases.
dy
5 3y 2 60; y(0) 5 30
dt
The results of the previous example and Practice Problem 2 can be summarised as:
●
if m < 0 then y(t) converges;
●
if m > 0 then y(t) diverges.
We say that an economic model is stable whenever the variables converge as t increases.
The above results indicate that an economic system represented by
dy
5 my 1 c
dt
Section 9.2 Differential equations 649
is stable if the coefficient of y is negative and unstable if it is positive. Of course, it could happen that m is zero. The differential equation then becomes
dy
5c
dt
which can be integrated directly to get
#
y(t) 5 cdt 5 ct 1 d
for some arbitrary constant d. The corresponding model is therefore unstable unless c is also
zero, in which case y(t) takes the constant value of d for all t.
We now investigate two applications of differential equations taken from macroeconomics
and microeconomics, respectively:
●
national income determination;
●
supply and demand analysis.
We consider each of these in turn.
9.2.1 National income determination
The defining equations of the usual two-sector model are
Y=C+I
(1)
C = aY + b
(2)
I = I*
(3)
The first of these is simply a statement that the economy is already in balance. The lefthand side of equation (1) is the flow of money from firms to households given as payment for
the factors of production. The right-hand side is the total flow of money received by firms,
either in the form of investment or as payment for goods bought by households. In practice,
the equilibrium values are not immediately attained, and we need to make an alternative
assumption about how national income varies with time. It seems reasonable to suppose that
the rate of change of Y is proportional to the excess expenditure, C + I − Y: that is,
dY
5 a(C 1 I 2 Y)
dt
(1′)
for some positive adjustment coefficient, a. This makes sense because
●
●
●
if C + I > Y, it gives dY/dt > 0, and so Y rises in order to achieve a balance between
expenditure and income
if C + I = Y, it gives dY/dt = 0, and so Y is held constant at the equilibrium level
if C + I < Y, it gives dY/dt < 0, and so Y falls in order to achieve a balance between
expenditure and income.
The usual relations (2) and (3) can be substituted into the new equation (1′) to obtain
dY
5 a(aY 1 b 1 I* 2 Y)
dt
5 a(a 2 1)Y 1 a(b 1 I*)
which we recognise as a differential equation of the standard form given in this section.
650 Chapter 9 Dynamics
Example
Consider the two-sector model
dY
5 0.5(C 1 I 2 Y)
dt
C 5 0.8Y 1 400
I 5 600
Find an expression for Y(t) when Y(0) = 7000. Is this system stable or unstable?
Solution
Substituting the expressions for C and I into
dY
5 0.5(C 1 I 2 Y )
dt
gives
dY
5 0.5(0.8Y 1 400 1 600 2 Y)
dt
5 20.1Y 1 500
The complementary function is given by
CF = Ae−0.1t
and for a particular solution we try
Y(t) = D
for some constant, D. Substituting this into the differential equation gives
0 = −0.1D + 500
which has solution D = 5000. The general solution is therefore
Y(t) = Ae−0.1t + 5000
The initial condition
Y(0) = 7000
gives
A + 5000 = 7000
and so A is 2000. The solution is
Y(t) = 2000e−0.1t + 5000
The first term is a negative exponential, so it converges to zero as t increases.
Consequently, Y(t) eventually settles down to an equilibrium value of 5000 and the system
is stable.
Section 9.2 Differential equations 651
Practice Problem
3. Consider the two-sector model
dY
5 0.1(C 1 I 2 Y)
dt
C 5 0.9Y 1 100
I 5 300
Find an expression for Y(t) when Y(0) = 2000. Is this system stable or unstable?
In the previous example and again in Practice Problem 3, we noted that the macroeconomic system is stable. If we return to the general equation
dY
5 a(a 2 1)Y 1 a(b 1 I*)
dt
it is easy to see that this is always the case for the simple two-sector model, since the coefficient of Y is negative. This follows because, as previously stated, a > 0, and because the
marginal propensity to consume, a, is less than 1.
9.2.2 Supply and demand analysis
The equations defining the usual linear single-commodity market model are
QS = aP − b
(1)
QD = −cP + d
(2)
for some positive constants a, b, c and d. As in Section 9.1, we have written Q in terms of P
for convenience. Previously, we have calculated the equilibrium price and quantity simply by
equating supply and demand: that is, by putting
QS = QD
In writing down this relation, we are implicitly assuming that equilibrium is immediately
attained and, in doing so, we fail to take into account the way in which this is achieved. A
reasonable assumption to make is that the rate of change of price is proportional to excess
demand, QD − QS: that is,
dP
5 a(QD 2 QS)
dt
for some positive adjustment coefficient, a. This makes sense because
●
●
●
if QD > QS, it gives dP/dt > 0, and so P increases in order to achieve a balance between supply and demand;
if QS = QD, it gives dP/dt = 0, and so P is held constant at the equilibrium level;
if QD < QS, it gives dP/dt < 0, and so P decreases in order to achieve a balance between
supply and demand.
Substituting equations (1) and (2) into equation (3) gives
dP
5 a[(2cP 1 d) 2 (aP 2 b)] 5 2a(a 1 c)P 1 a(d 1 b)
dt
which is a differential equation of the standard form.
652 Chapter 9 Dynamics
Example
Consider the market model
QS 5 3P 2 4
QD 5 25P 1 20
dP
5 0.2(QD 2 QS)
dt
Find expressions for P(t), QS(t) and QD(t) when P(0) = 2. Is this system stable or unstable?
Solution
Substituting the expressions for QD and QS into
dP
5 0.2(QD 2 QS)
dt
gives
dP
5 0.2[(25P 1 20) 2 (3P 2 4)] 5 21.6P 1 4.8
dt
The complementary function is given by
CF = Ae−1.6t
and for a particular solution we try
P(t) = D
for some constant D. Substituting this into the differential equation gives
0 = −1.6D + 4.8
which has solution D = 3. The general solution is therefore
P(t) = Ae−1.6t + 3
The initial condition
P(0) = 2
gives
A+3=2
and so A is −1. The solution is
P(t) = −e−1.6t + 3
Corresponding expressions for QS(t) and QD(t) can be found from the supply and demand
equations, which give
QS(t) = 3P − 4 = 3(−e−1.6t + 3) − 4 = −3e−1.6t + 5
QD(t) = −5P + 20 = −5(−e−1.6t + 3) + 20 = 5e−1.6t + 5
Note that all three expressions involve a negative exponential that converges to zero as t
increases, so the system is stable. The price P(t) eventually settles down to the equilibrium
price, 3, and QS(t) and QD(t) both approach the equilibrium quantity, 5.
Section 9.2 Differential equations 653
Practice Problem
4. Consider the market model
QS 5 2P 2 2
QD 5 2P 1 4
dP 1
5 (QD 2 QS)
dt
3
Find expressions for P(t), QS(t) and QD(t) when P(0) = 1. Is this system stable or
unstable?
In the previous example and again in Practice Problem 4, we noted that the single commodity
market model is stable. If we return again to the general equation
dP
5 2a(a 1 c)P 1 a(d 1 b)
dt
it is easy to see that this is always the case, since the coefficient of P is negative. This follows
because, as previously stated, a, a and c are all positive.
Key Terms
Adjustment coefficient The constant of proportionality in the simple macroeconomic model,
in which the rate of change of national income is assumed to be proportional to excess expenditure.
Arbitrary constant A letter representing an unspecified constant in the general solution of a
differential equation.
Complementary function of a differential equation The solution of the differential
dy
5 my 1 c when the constant c is replaced by zero.
equation
dt
Differential equation An equation connecting derivatives of an unknown function.
Equilibrium value of a differential equation A solution of a differential equation that does not
vary over time; it is the limiting value of y(t) as t tends to infinity.
General solution of a differential equation The solution of a differential equation that contains
an arbitrary constant. It is the sum of the complementary function and a particular solution.
Initial condition The value of y(0) that needs to be specified to obtain a unique solution of a
differential equation.
Particular solution of a differential equation Any one solution of a differential equation
dy
5 my 1 c.
such as
dt
Stable equilibrium An economic model in which the solution of the associated differential
equation converges.
654 Chapter 9 Dynamics
Exercise 9.2
1. Use integration to solve each of the following differential equations subject to the given
initial conditions:
(a)
dy
5 2t; y(0) 5 7
dt
(b)
dy
5 e23t; y(0) 5 0
dt
(c)
dy
5 t 2 1 3t 25; y(0) 5 1
dt
2. Solve the differential equation
dy
5 23y 1 180
dt
in the case when the initial condition is
(a) y(0) = 40 (b) y(0) = 80 (c) y(0) = 60
Comment on the qualitative behaviour of the solution in each case.
3. A principal of $60 is invested. The value, I(t), of the investment, t days later, satisfies
the differential equation
dI
5 0.002I 1 5
dt
Find the value of the investment after 27 days, correct to 2 decimal places.
4. Consider the two-sector model
dY
5 0.5(C 1 I 2 Y )
dt
C 5 0.7Y 1 500
I 5 0.2Y 1 500
Find an expression for Y(t) when Y(0) = 15 000. Is the system stable or unstable?
5. Consider the two-sector model
dY
5 0.3(C 1 I 2 Y )
dt
C 5 0.8Y 1 300
I 5 0.7Y 1 600
Find an expression for Y(t) when Y(0) = 200. Is this system stable or unstable?
6. Consider the market model
QS 5 3P 2 1
QD 5 22P 1 9
dP
5 0.5(QD 2 QS)
dt
Find expressions for P(t), QS(t) and QD(t) when P(0) = 1. Is this system stable or
unstable?
Section 9.2 Differential equations 655
7. Oil reserves decrease at a constant proportional rate, k > 0, so that
dN
5 2kN
dt
where N denotes the number of barrels of oil remaining in t years’ time.
(a) Write down an expression for N(t) when current oil reserves are A barrels.
(b) Show that the time taken for oil reserves to fall to half their current levels is (ln 2)/k.
Exercise 9.2*
1. Find the solution of the differential equation
dy
4
5 3t 2 2
dt
t
with initial condition, y(0) = 4.
2. Consider the market model
QS 5 4P 2 3
QD 5 22P 1 13
dP
5 0.4(QD 2 QS)
dt
Find an expression for QD(t) when P(0) = 2.
3. A principal of $4000 is invested at an annual interest rate of 6%, and the future value
of this investment t years later is S(t), which satisfies
dS
5 0.06S
dt
(a) Solve this equation to express S in terms of t.
(b) What type of compounding is represented by this model?
4. Solve the differential equation
dy
5 8e22t
dt
with initial condition y(0) = 10.
Sketch the graph of y against t.
5. Consider the two-sector model
dY
5 0.4(C 1 I 2Y)
dt
C 5 0.6Y 1 400
I 5 0.8Y 1 500
Given that Y(0) = 100, find the value, correct to the nearest whole number, of Y(2.4).
➜
656 Chapter 9 Dynamics
6. Consider the two-sector macroeconomic model
dY
5 0.2(C 1 I 2Y)
dt
C 5 0.8Y 1 420
I 5 300
(a) Find an expression for Y(t) when Y(0) = 8000.
(b) Hence find an expression for the savings function, S(t).
(c) Find the time taken for income to fall to 4150, and find the rate of change of income
at this time. Give your answers to the nearest whole number.
7. The value of an economic variable over time satisfies the relation
R(t) 5
6
1 3e20.4t
1 1 t2
Find the equilibrium value of R.
8. A simple model of the growth of an economy is based on three assumptions.
(1) Savings, S, are proportional to income, Y, so that
S = aY
(a > 0)
(2) Investment, I, is proportional to the rate of change of Y so that
I5b
dy
dt
(b . 0)
(3) Investment and savings are equal so that
I=S
Use these assumptions to show that
a
dY
5 Y
b
dt
and hence write down a formula for Y(t) in terms of Y(0). Is this system stable or
unstable?
9. A model predicts that monthly sales, S(t), of a new product after t months satisfy the
differential equation
dS
5 k(M 2 S) with initial condition S(0) = C
dt
where k, M and C are positive constants.
(a) Show that S(t) = M(1 − e−kt ) + Ce−kt.
(b) Sketch a graph of S against t and comment on the forecast for sales in the long run.
(c) If S(0) = 100, S(6) = 2000 and M = 5000, what does the model predict that monthly
sales will be after one year of trading?
Section 9.2 Differential equations 657
10. Show, by substituting into the differential equation, that
y(t) 5 Aemt 2
c
m
is a solution of
dy
5 my 1 c
dt
11. Consider the differential equation
dy
5 22y 1 5e3t
dt
(a) Find the complementary function.
(b) By substituting y = De3t into this equation, find a particular solution.
(c) Use your answers to parts (a) and (b) to write down the general solution and hence
find the specific solution that satisfies the initial condition, y(0) = 7.
(d) Is the solution in part (c) stable or unstable?
12. Consider the differential equation
dy
5 2y 1 4t 2 3
dt
(a) Find the complementary function.
(b) By substituting y = Dt + E into this equation, find a particular solution.
(c) Use your answers to parts (a) and (b) to write down the general solution and hence
find the specific solution that satisfies the initial condition, y(0) = 1.
(d) Is the solution in part (c) stable or unstable?
Formal mathematics
We conclude this chapter with a formal derivation of the solution of the general difference
equation considered in Section 9.1:
Yt = bYt−1 + c with Y0 = a.
Setting t = 1, 2 and 3 in turn gives
Y1 = bY0 + c = ab + c
Y2 = bY1 + c = b(ab + c) + c = ab2 + bc + c
Y3 = bY2 + c = b(ab2 + bc + c) + c = ab3 + b2c + bc + c
Continuing gives
Yt = abt + bt−1c + bt−2c + . . . + b2c + bc + c
= abt + c(1 + b + b2 + . . . + bt−1)
The expression in the brackets is the first t terms of a geometric series with ratio b. If we
assume that b ≠ 1, we can write
Yt 5 abt 1 c
5 a1
bt 2 1
b21
c
c
bt 2
b21
b21
which confirms the ‘Yt = Abt + PS’ form used in Section 9.1.
Example
Solve the difference equation Yt = −2Yt−1 + 9; Y0 = 4
Solution
Substituting a = 4, b = −2 and c = 9 into the general solution gives
Yt 5 4 1
9
9
(22)t 2
5 (22)t 1 3
22 2 1
22 2 1
[This is the same difference equation considered in Practice Problem 2 (b) in Section 9.1.]
Multiple choice questions
Question 1
The variation of capital stock, K, satisfies the differential equation
dK
5 3 t 1 2 where
dt
K = 25 when t = 1. Work out the capital stock when t = 4.
A. 40
B. 20
C. 25
D. 30
E. 45
Question 2
Consider the two-sector model:
Yt = Ct + 400
Ct = 0.8Yt−1 + 100
If Y0 = 450, work out the value of C2.
A. 860
B. 788
C. 708
D. 786
E. 888
Question 3
Solve the differential equation
dy
5 0.6y with initial condition y(0) = 4 to find y(6) correct
dt
to the nearest whole number.
A. 2
B. 3
C. 146
D. 72
E. 36
Question 4
Find the complementary function of the difference equation, Yt − 3Yt−1 = 18
A. A(−3)t
B. −2
C. −9
D. A(3t)
E. −9 + A(−3)t
660 Chapter 9 Dynamics
Question 5
Find the particular solution of the differential equation,
dy
t
5 23y 1 6e
dt
A. 6e−3t
B. 1.5et
C. Ae−3t
D. 6et
E. Ae−3t + 6et
Question 6
Describe the qualitative behaviour of the solution of the difference equation
3
Yt 5 2 Yt21; Y0 5 10
4
A. Oscillatory convergence
B. Oscillatory divergence
C. Uniform convergence
D. Unstable convergence
E. Stable divergence
Question 7
Find the solution of the differential equation
dy
1 4y 5 200 with initial condition, y(0) = 30
dt
A. 50 + 20e4t
B. 50 − 20e−4t
C. −50 − 20e−4t
D. −50 −20e4t
E. 50 + 20e−4t
Question 8
Consider the supply and demand equations
QSt = 2Pt−1 −10
QDt = −4Pt + 40
Assuming that the market is in equilibrium, find an expression for Qt when P0 = 4.
A. 2
8 1 t 20
1
2
3 2
3
13 1 t 25
B. 2
1
2
3
2
3
Multiple choice questions 661
C.
52 1 t 20
1
2
3
2
3
13 1 t 25
D. 2
1
3 2
3
E. 2
8 1 t 20
1
3 2
3
Question 9
Consider the two-sector model:
dY
5 0.6(C 1 I 2 Y)
dt
C = 0.9Y + 600
I = 400
Find an expression for Y(t) when Y(0) = 9000.
A. 8000e−0.06t + 10000
B. 10000 − 1000e−0.06t
C. 8000e−0.6t + 1000
D. 1000e−0.1t + 1000
E. 10000 − 8000e−0.6t
Question 10
Consider the general supply and demand model:
QS 5 aP 2 b
QD 5 2cP 1 d
dP
5 a(QD 2 QS )
dt
where a, b, c, d and a are all positive constants.
Find an expression for P(t) when P(0) = b.
2a(a1c)t
1b1d
A. (a 1 c 1 b(b 1 d))e
a1c
(b 1 d) (e2a(a1c)t 21) 1 b(a 1 c)e2a(a1c)t
B.
a1c
2a(a1c)t
1b1d
C. (b 1 d 1 b(a 1 c))e
a1c
2a(a1c)t
2 (b 1 d)
D. (b 1 d 2 b(a 1 c))e
a1c
2a(a1c)t
) 1 b(a 1 c)e2a(a1c)t
E. (b 1 d) (1 2 e
a1c
Examination questions
Question 1
Find the equilibrium price and quantity given the following static supply and demand
equations:
QS = 6P − 10
QD = −4P + 45
The equations for the corresponding dynamic model are:
QSt = 6Pt−1 −10
QDt = −4Pt + 45
Find an expression for Pt in terms of t given the initial condition, P0 = 8 and describe the
behaviour of Pt as t increases.
Question 2
(a) Solve the differential equation
dy
4
1 6x 2 given the condition, y(4) = 3.
5
dx
x
(b) The population, N, of a small country is currently 5 million. It is expected that this
population will grow according to the differential equation.
dN
5 0.015N
dt
where t is measured in years.
Solve this equation to express N in terms of t and hence predict how many years it
will take for the population to double. Give your answer to one decimal place.
Question 3
Government expenditure, G, satisfies the differential equation
dG
5 20.05G
dt
where G is measured in billions of dollars and time t is in years. The initial condition is
G(0) = 500.
(a) Solve this equation to express G in terms of t.
(b) Find the value of (i) G when t = 6; (ii) t when G = 125. Round your answers to three
significant figures.
Examination questions 663
Question 4
A two sector lagged-income model is defined by
Yt = Ct + It
Ct = 0.8Yt−1 + 200
It = 1000
where Yt, Ct and It denote national income, consumption and planned investment in time
period t. The initial value of national income is Y0 = 5000.
(a) Use these relations to work out Y2 and C2.
(b) Derive a first-order difference equation relating Yt and Yt−1.
(c) Solve the difference equation to express Yt in terms of t.
(d) Verify that your general formula in part (c) agrees with the value of Y2 calculated in
part (a).
(e) Find an expression for Ct in terms of t and hence determine the time period in which
the consumption first exceeds 4750.
Question 5
National income Y varies over time t according to the following model:
dY
5 0.6(C 1 I 2 Y)
dt
C 5 0.8Y 1 600
I 5 800
where C is consumption and I is planned investment. Initially, Y = 2000.
(a) Find an expression for Y(t) and deduce that the system is stable.
(b) Sketch a graph of Y against t.
(c) Find the time taken for national income to rise to 90% of its equilibrium value. Round
your answer to one decimal place.
(d) Find an expression for C(t).
Question 6
Consider the supply and demand equations:
QSt = 2Pt + aPt−1 −20
QDt = 100 − 8Pt
for some parameter, a.
Assuming that the market is in equilibrium, show that Pt = 12 − 0.1aPt−1.
If P0 = 10, find an expression for Pt in terms of t and a.
If the solution displays oscillatory convergence, state the range of values of a and find the
particular value of a if P1 = 9.5.
Answers
Chapter 1
8 (a) 16
(b) Presented with the calculation, −42, your
Section 1.1
(d) 5
(e) 36
(f) −1
calculator uses BIDMAS, so squares first to get
16 and then subtracts from zero to give a final
answer, −16. To obtain the correct answer you
need to use brackets:
2 (a) −1
(b) −7
(c) 5
(
(d) 0
(e) −91
(f) −5
3 (a) 19
(b) 1500
(c) 32
4 (a) x + 9y
(b) 2y + 4z
(c) not possible
Practice Problems
1 (a) −30
(b) 2
(c) −5
(d) 8r2 + s + rs − 3s2
(e) −4f
(f) not possible
(g) 0
5 (a) 5z − 2z
2
(b) −3y
(d) 35
(c) z − x
(d) 5Q(1 − 2Q)
(b) x2 − y2
(b) 21; no
10 (a) 43.96
(b) 1.13
(c) 10.34
(e) 27.38
(f) 3.72
(g) 62.70
(e) 3x + 3y + 3z
(f) 3x2 − 4x
12 (a) 5(5c + 6)
(d) 4(4x − 3y)
(b) 9(x − 2)
(c) x(x + 2)
(e) 2x(2x − 3y)
(f) 5(2d − 3e + 10)
2
(d) 6s2 + 23s + 21 (e) 2y2 + 5y + 3
(g) 9n2 − 4
8 (a) (x + 8) (x − 8)
Exercise 1.1
(b) 3
(c) −4
(d) 1
(e) −12
(f) 50
(g) −5
(h) 3
(i) 30
( j) 4
(b) −3
(c) −11
(d) 16
(e) −1
(f) −13
(g) 11
(h) 0
(i) −31
( j) −2
(d) 6xyz + 2xy
(e) 10a − 2b
(g) 11 − 3p
(h) x + 10x
15 (a) (x + 2)(x − 2) (b) (Q + 7)(Q − 7)
(c) (x + y)(x − y) (d) (3x + 10y)(3x − 10y)
16 (a) 4x2 + 8x − 2 (b) −13x
17 S = 1.2N + 3000E + 1000(A − 21);
18 (a) C = 80 + 60L + K
(c) H = 5a + 10b
(d) X = Cd + cm
(e) −41
(f) −3
(g) 18
(h) −6
(i) −25
( j) −6
1 (a) 3
4 (a) 2PQ
(b) 8I
(c) 3xy
2 (a) 2 − 7 − (9 + 3) = −17
(d) 4qwz
(e) b2
(f) 3k2
(e) 4c − 3cd (f) 2st + s + t + 9
2
(e) 70
7 (a) 1
(e) −30
(b) 5
(f) 44
2
(c) 2000
(c) −6
Exercise 1.1*
(b) 5
(c) −7
(b) 8 − (2 + 3) − 4 = −1
(b) 4x − 7y (c) 9a + 2b − 2c
(d) 96
$204 000
(b) C = 10 + 1.25x
(d) −15
(b) 18
(f) 17x + 22y
2
(c) 18
(d) x + 2x
(f) 10t 2 − 31t − 14
(b) 11x2 − 3x − 3 (c) 14xy + 2x
(b) 2
6 (a) 10
(c) d 2 − 5d − 24
(h) a2 − 2ab + b2
14 (a) 6x + 2y
(b) (2x + 9) (2x − 9)
2
(h) 2.39
(c) 4x + 12
13 (a) x + 7x + 10 (b) a + 3a − 4
(d) 5x2 − 3xy + 5x − 2y2 + 2y
5 (a) 19w
(d) 0.17
(b) 15x − 6y
2
(c) x2 + 2xy + y2
3 (a) −3
5
9 (a) 9
(g) −2x − 5y + 4z
(c) 3(2x − y + 3z)
2 (a) −1
x2
)
(d) 21x − 7
2
(b) 4(4w − 5q)
1 (a) −20
4
11 (a) 7x − 7y
6 (a) 7(d + 3)
7 (a) x2 + x − 6
2
(c) 7 − (2 − 6 + 10) = 1
3 (a) −6
(e) −1
(d) −6
4 (a) 6
(b) 6
(c) −5
(d) −96
(f) 6
5
(g)
4
(h) 63
(b) 2
(c) 5
5 −y + xy − 5x + 2y − 6
2
ANSWERS 665
6 (a) 2x − 2y (b) 2x
(c) −2x + 3y
7 (a) x − 2x − 24
(b) 6x − 29x + 35
2
4 (a)
2
(c) 6x + 2xy − 4x
(d) 12 − 2g + 3h − 2g2 + gh
2
(d)
(e) 2x − 2x2 − 3xy + y − y2 (f) a2 − b2 − c2 − 2bc
8 (a) 3(3x − 4y)
(b) x(x − 6)
5
(c) 5x(2y + 3x)
(d) 3xy(y − 2x + 4)
(e) x (x − 2)
(f) 5xy3(12x3y3 − 3xy + 4)
2
9 (a) ( p + 5)(p − 5)
(b) (3c + 8)(3c − 8)
(c) 2(4v + 5d)(4v − 5d)
10 (a) 112 600 000
(d) (4x2 + y2)(2x + y)(2x − y)
(b) 1.7999
(c) 283 400
(d) 246 913 577
11 (a) π = 12y − 3x − 800
(c) 0 ≤ y ≤ x
(b) $5800
(b) (L − 0.2K)(L + 0.2K)
Section 1.2
Practice Problems
3
5
4
5
2 (a)
3
8
(b)
3 (a)
2
7
4 (a)
(c)
1
2y
(d)
1
2 1 3x
(e)
1
x24
7
4
(c)
3
4
(d)
1
18
(b)
11
15
(c)
5
36
5
x12
(b)
x(x 1 1)
x 1 10
5
x 11
(d)
x 13
(x 1 1)(x 1 2)
5 (a) 6
(b)
(c)
(b) 12
(c) 2
3
4
6 (a) 12 > 9 (true)
9
5
3
7
(b) 2
(c)
5
6
(d)
7
20
(e)
7
18
(f)
5
6
(g)
5
8
(h)
2
5
(i)
7
12
( j)
1
30
(k)
2
27
(l)
21
1
5 10
2
2
1
x
(b)
2
5
(c)
3x 2 2
x2
(d)
7y 1 2x
(e) 3
xy
(f)
15c 1 10d
36
(g)
x12
x 13
(i)
t
20
( j) 1
(b) 6
(c) 18
(d) 2
(e) 10
(f) −1
(g) 60
(h) 1
(i) −5
( j) −3
(k) −2
(l) 23
(n) 3
(o)
6 (a)
8 (a)
9 (a) 5
1
4
11 (a) x > 1
(e) 2
3
2
(b) 12 > 6 (true)
13 (a) 2
7
26
(e) 2 > 1 (true)
(f) −24 > −12 (false)
Exercise 1.2*
(g) −6 > −3 (false)
(h) −2 > −1 (false)
1 (a)
(i) −4 > −7 (true)
(b) x ≥ 2
3
(b)
4
3
(c)
5
7 14
2 (a)
; 20 25
3
(b) 1
5
2x
1
3 (a)
(b)
3
2x
1
(c)
ac
1
(d)
3
4
1
(e)
51
3
3
(g)
2 (a)
2
(d)
3xy
x23
2
18h 2
7
2
3
2
3
1
4
3a
(e)
4b
(b) x ≤ 3
(c) x < −3
(d) x > 2
(e)
(b) x ≤ 10
(b)
3
2x 2 1
(c) 4
(e)
1
x 26
(f)
x13
x14
1
2x2 2 5x 1 3
(h)
2x 1 5y
3
5
7
(b)
1
10
(c)
3
2
(d)
5
48
22
13
(f)
11
9
(g)
141
35
(h)
34
5
( j)
17
10
(k)
7
9
(l) 4
(d) −1
Exercise 1.2
1
1 (a)
2
(h)
2
12 3
x
(d) same as (c)
1
3
10 (a), (d), (e), (f)
(c) 3 > 0 (true)
7 (a) x > −7
1
x22
factors on top and bottom.
(m) 3
(d)
(e)
b
2a 1 1
x 21
x 21
1
5
5 ; other two have no common
2x 2 2
2(x 2 1) 2
(c) (K + L)2
1 (a)
2
32e
(c)
7 38
(d) π = 9x − 800
12 (a) 2KL(L + 2)
1
p
(b)
x24
2q 1 3r
(i) 6
666 ANSWERS
3 (a) x + 6
(b)
5
(c)
xy
2
x 11
1
or equivalently 1 1
x
x
5x 1 2
(d)
6
x2 1 x 2 2
(g)
x11
11
4 (a) 2
(b) 1
7
3x 1 5
7x 1 3
(e)
(f)
x2
x(x 1 1)
x13
(h)
x(x 1 1)
35
(c) 2
(d) 8
9
(f)
1
4
(g) 2
(i) 9
( j)
71
21
(k) 7
(l) −9
(m) 1
(n) −5
(o) 3
(p) 5
5 1.6 1
11
7
Check
(−1, 2)
2(−1) + 3(2) = −2 + 6 = 4
✓
(−4, 4)
2(−4) + 3(4) = −8 + 12 = 4
✓
(5, −2)
(2, 0)
2(5) + 3(−2) = 10 − 6 = 4
✓
2(2) + 3(0) = 4 + 0 = 4
✓
The graph is sketched in Figure S1.2.
The graph shows that (3, −1) does not lie on the line.
4
5
(e)
Point
(h) 8
y
5
4
(24, 4)
3
5x
5 6.75; $7.21
7
6 (a) $3221.02 (b) $60 000
7 (a) x , 28
(21, 2)
1
(c) 10
3
12
(b) x . 2
(c) x ≤ 13
5
13
(d) x > −3
2
25 24 23 22 21
21
(2, 0)
1 2 3
(3, 21)
4
(5, 22)
22
(e) −1 < x ≤ 3
this point does
not lie on line
23
8 (a) 9000 + 50x
x
5
24
(b) 10 800 ≤ 9000 + 50x ≤ 12 500;
36 ≤ x ≤ 70
25
9 −3, −2, −1, 0
10 (a)
11
3
2x 2 1
(b) −1
(c) x $
11
5
Figure S1.2
3 (2, 1) and (−2, −5) lie on the line.
x(x 2 1)
2
The line is sketched in Figure S1.3.
Section 1.3
Practice Problems
y
1 From Figure S1.1, note that all five points lie on a
5
4
straight line.
3
2
y
1
5
(2, 5)
25 24 23 22 21
21
4
3
22
(1, 3)
2
1
22
(22, 23)
23
24
(23, 25)
Figure S1.1
25
1
2
3
24
(0, 1)
1
2
3
4
5
x
4
3x 2 2y 5 4
23
(22, 25)
25 24 23 22 21
21
(2, 1)
25
Figure S1.3
4 (0, −1) and (2, 0) lie on the line.
The graph is sketched in Figure S1.4.
5
x
ANSWERS 667
y
y
1
12
5
4
1
3
x 2 2y 5 2
2
1
(2, 0)
1
25 24 23 22 21
21
21
2
3
1
2
intercept
4
x
5
1
22
2
1
2
1
2
1
1
x
12
1 unit
(0, 21)
21
22
23
1
2 units
212
24
22
25
1
22 2
Figure S1.4
y 5 12 2 2x
23
5 From Figure S1.5, the point of intersection is (1, −1/2).
Figure S1.7
y
Exercise 1.3
3x 2 2y 5 4
5
1 From Figure S1.8, the point of intersection is (2, 3).
4
x 2 2y 5 2
3
2
y
1
1
2
25 24 23 22 21
21
3
4
9
Q
x
5
7
point of
intersection
(1, 21 )
22
23
R
8
6
2
5
24
4
25
(2, 3)
3
2
Figure S1.5
1
6 (a) a = 1, b = 2. The graph is sketched in Figure S1.6.
25 24 23 22 21
(b) a = −2, b = 1/2. The graph is sketched in Figure S1.7.
S
y
3
intercept
22
1
21
21
22
Figure S1.6
1 unit
1
y5x1 2
23
2
Figure S1.8
1 unit
2
3
x
P
1
21
22
2
3
4
5
x
668 ANSWERS
2 (a) The graph is sketched in Figure S1.9.
y
3
Cost in $
1 unit
1000
900
800
700
600
500
400
300
200
100
23
22
2
1 unit
1
2
1
21
3
x
21
22
y 5 2x
23
1000 2000 3000 4000 5000 Distance in km
Figure S1.11
Figure S1.9
y
(b) (i) $540 (ii) 2500 km
5
3 A, C, D, E
(b) −1;
4 (a) 6
4
(6, 2), (1, −1)
3
5 x
y
2
0
8
1
6
0
25 24 23 22 21
21
3
4
1
2
3
4
The graph is sketched in Figure S1.10.
23
6
1 unit
22
y 5 12 x 2 3
5
2 units
24
25
y
26
9
8
Figure S1.12
7
9 (a) C = 4 + 2.5x
(b) The graph is sketched in Figure S1.13.
6
5
Cost in $
4
3
50
4x 1 3y 5 24
2
40
1
1
2
3
4
5
6
7
30
x
20
Figure S1.10
10
6 (a) (−2, −2) (b) (2, 11/2)
(b) 3, −1
7 (a) 5, 9
(e) 22,
5
2
(c) (11/2, 1)
(d) (10, −9)
(c) −1, 13
(d) 1, 4
5
(f) 5, −6
8 (a) The graph is sketched in Figure S1.11.
(b) The graph is sketched in Figure S1.12.
Figure S1.13
(c) 8 miles
10
15
Distance in miles
20
7 x
ANSWERS 669
10 (a) The graph is sketched in Figure S1.14.
2 (a) There are no solutions.
(b) There are infinitely many solutions.
3 x = 1, y = −1, z = 1
N
350
Exercise 1.4
300
1 (a) x = −2, y = −2
(b) x = 2, y = 3/2
260
250
(c) x = 3/2, y = 1
(d) x = 10, y = −9
2 (a) x + y = 3500
200
190
30x + 25y = 97 500
150
(b) 1500
3 The lines are sketched in Figure S1.15.
100
(a) Infinitely many
50
(b) No solution
5
7
10
14 15
20 n
y
Figure S1.14
(b) (i) N = 260
(ii) n = 7
2
(c) Slope = 10; this is the number of staff employed
in each café. Intercept = 120; number of staff
employed managing the company.
11 (a) $9000
(b) $18 600
(c) $500
(d) $12
24 23 22 21
21
2 (a) (2, 5)
(b) (1, 4)
(c) (−2, 3)
3 (a) 7, −34
(b) −1, 1
(c)
3
, 23
2
(d) 2,
3
4
5
6
x
22
25
5
2
Figure S1.15
(f) 0, 2
not intercept the y-axis.
4 (b) and (d)
4 (a) Infinitely many
(b) No solution
5 k = −1
5 (a) x + 100 (b) 3.5x + 35
(c) 26
6 (a) −12
(c) 300
(b) 408
(d) $20
a
d
and 2 , respectively, so the
b
e
a
d
lines are parallel when 5 which gives ae = bd.
b
e
7 (1) Gradients are 2
(2) Parallel lines so no solution.
0,
2
24
(d) (−8, 3)
(g) The vertical line, x = 4, has no gradient and does
8
1
23
1 (5, −2), (10, 1), (0, −5)
1
, 0
5
2x 2 3y 5 6
1
x 2 32 y 5 3
Exercise 1.3*
(e)
4x 2 6y 5 18
3
c
c
,
,0
b
a
1 (a) x = 2, y = 5
(c) x = −2, y = 3
2 (a) a = 4, b = 8
(b) x = 1, y = 4
(d) x = −8, y = 3
(b) a = −3, b Þ
subtract to eliminate x.
(b) x = −1, y = 3, z = 4
5 (a) No solution.
(b) Infinitely many solutions.
Practice Problems
1
2
(b) x = 3, y = 2
1
2
3 Multiply the first equation by d, the second by a and
4 (a) x = 3, y = −2, z = −1
Section 1.4
1 (a) x 5 1, y 5 2
Exercise 1.4*
6 k = 6; no solutions otherwise.
7 $98.85
670 ANSWERS
Section 1.5
P
Practice Problems
1 (a) 0
(b) 48
(c) 16
(d) 25
(e) 1
11
10
(f) 17
The function g reverses the effect of f and takes you
back to where you started. For example, when 25 is
put into the function f, the outgoing number is 0; and
when 0 is put into g, the original number, 25, is
produced. We describe this by saying that g is the
inverse of f (and vice versa).
7
P 5 13 Q 1 7
2 The demand curve that passes through (0, 75) and
(25, 0) is sketched in Figure S1.16. From this diagram
we see that
(a) P = 6 when Q = 23
9
12
Q
Figure S1.17
(b) Q = 19 when P = 18
3 (a) Demand is 173. Additional advertising
expenditure is 12.
(b) Normal
P
4 (a) 23
75
(b) Substitutable; a rise in PA leads to an increase in Q.
P 5 23Q 1 75
(c) 6
5 a = −6, b = 720
6 (a) 20, 10, 45; line passes through these three points.
(b) Line passing through (50, 0) and (0, 50)
Q = 20, P = 30
(c) Price increases; quantity increases.
18
7 6
6
8 P1 = 40, P2 = 10; Q1 = 30, Q2 = 55
19 23 25
Q
9 (a) Q = 30
(b) Substitutable; e.g. since coefficient of Pr is positive.
Figure S1.16
(c) P = 14
(d) (i) slope = −20, intercept = 135
3 (a) Q = 21 and P = 36
(b) Q = 18 and P = 48. The consumer pays an
additional $12. The remaining $1 of the tax is
paid by the firm.
4 P1 = 4, P2 = 7, Q1 = 13, Q2 = 14.
The goods are complementary.
Exercise 1.5
1 (a) 21
(b) 45
(c) 15
(d) 2
(e) 10
(f) 0; inverse
2 The supply curve is sketched in Figure S1.17.
(a) 11
(b) 9
(c) 0; once the price falls below 7, the firm does not
plan to produce any goods.
(ii) slope = 2
1
, intercept = 6.75
20
Exercise 1.5*
1 (a) As PS rises, consumers are likely to switch to
the good under consideration, so demand
for this good also rises: that is, the graph shifts
to the right.
(b) As PC rises, demand for the bundle of goods
as a whole is likely to fall, so the graph shifts
to the left.
(c) Assuming that advertising promotes the good
and is successful, demand rises and the graph
shifts to the right. For some goods, such as drugs,
advertising campaigns are intended to discourage
consumption, so the graph shifts to the left.
ANSWERS 671
3
2
2 m52 , c59
1
(y 1 6)
9
4 (a) x 5
3 0 and 30
(c) x = 2y
4 (1) P = 30, Q = 10
(2) New supply equation is 0.85P = 2QS + 10;
(e) x 5
P = 33.6, Q = 9.28.
5 (a) 53, 9
5 (a) P 5
(b) $324
6 P1 = 20, P2 = 5, P3 = 8; Q1 = 13, Q2 = 16, Q3 = 11
7 Change supply equation to P = 2QS + 40 + t.
In equilibrium
−5Q = −20 + t
1 4
17
3 y
Q
b
2
a
a
(b) Y 5
b1I
12a
(b) H 5
2DR
Q2
HQ2
2R
1 (1) (a) multiply by 9
(b) multiply by −1
(b) P = 45, Q = 5
(c) square
8 (a) a > 0, b > 0, c < 0, d > 0
(b) Q 5
(f) x 5
Exercise 1.6*
3
Substitute to get P 5 48 1 t.
5
(a) t = 5; firm pays $2
1
22
y
3
y12
7 (a) D 5
t
Q542
5
(e) square
Practice Problems
1 (a) Q = 8
(c) Q = 2 × 17 − 26 = 8
y
6
2 (a) x 5
3 (a) x 5
11 a
y
12 c
(b) x 5
1 1
1 1y
11 5
7 y
7y
(b) x 5
22 2 4y
y 21
Exercise 1.6
1 Q5
(b) y = 2(x + 5)
5
x2
3 (a) multiply by 5
(d) y = 2(x + 4)2 − 3
add 3
subtract 9
multiply by 4
4
28
y
add 7
(f) reciprocate
multiply by 2
reciprocate
c2a
b
(c) x = (g − e)2 − f
n2
1 m
m2
multiply by 4
(b) x = 3 − y
(d) x 5
y2 2 5
3
(b) x 5
a2 1 b
a11
(d) x 5
ma2
1n
b2
(f) x 5
a2 1 b2
b2a
V11
; 11
V25
4 r 5 100
n
S
21
P
(b) T 5
G 1 b 1 I 2 Y(1 2 a 1 at)
a
(c) t 5
G 1 b 1 I 2 Y 1 aY 2 aT
aY
(d) a 5
G1b1I2Y
T 2 Y 1 tY
subtract 6
(e) divide by 2
(g) add 3
add 5 square root
5 (a) G = Y(1 − a + at) + aT − b − I
multiply by 5
(c) multiply by 6
(d) square
(e) x 5 6
2 (a) x 5
3 t5
2 (a) y = 2x + 5
(b) add 3
y18
5
(e) x 5
1
P 2 4; 22
2
(c) y 5
(c) x 5 6
(multiply out brackets)
add 3
add 8 reciprocate
y21
(2) (a) x 5
9
Section 1.6
add 1
multiply by 5 subtract 8
(d) multiply by 3
d2b
ad 2 bc
and P 5
a2c
a2c
(b) Q = 2P − 26
(d) x = 5(y − 8)
1
b
2
aQ
a
(c) P 5
6 x5
−3Q + 60 = 2Q + 40 + t
(b) x = 3y – 4
2
672 ANSWERS
Section 1.7
Exercise 1.7*
Practice Problems
1 (a) S = 0.3Y − 30
1 S = 0.2Y − 25
2 Y = 210; Y = 215
2 C5
aI* 1 b
12 a
3 a5
Y 2 b 2 I*
Y
3 Y = 400
4 Y = 2950; r = 8
The IS and LM curves shown in Figure S1.18
confirm this, since the point of intersection has
coordinates (8, 2950). A change in I does not affect
the LM schedule. However, if the autonomous level
of investment increases from its current level of
1200, the IS curve moves upwards, causing both r
and Y to increase.
Y
4283 1
3
LM curve
2950
(b) S 5
10Y 2 500
Y 1 10
4 825
5 Y = 2500, r = 10
6 a5
Y 2 74
; a = 0.6, C = 131
Y 2 20
7 (a) increases
(b) decreases
Multiple Choice Questions
1 E
2 B
3 C
4 D
5 A
6 A
7 E
8 B
9 D
10 A
11 D
12 D
13 C
14 E
15 A
16 A
17 D
18 C
Examination Questions
1 (a) The graphs are sketched in Figure S1.19.
1350
IS curve
y
25
7
8
25 10
Deliver 4U
r
20
EatMeNow
Figure S1.18
15
Exercise 1.7
1 MPC = 0.75; MPS = 0.25
10
2 (a) MPC = 0.7; MPS = 0.3
5
(b) C = 0.7Y + 100
3 (a) 40
(b) 0.7; Y 5
10
(C 2 40); 100
7
4 (a) S = 0.1Y − 72
(b) S = 0.2Y − 100
5 (a) 325
(c) 100
(b) 225
6 10a + b = 28
30a + b = 44
a = 0.8, b = 20; Y = 165
7 187.5
5
Figure S1.19
10
15
20
25
x
ANSWERS 673
(b) 12 miles
(c) EatMeNow: y = 0.5x + 9, Deliver4U: y = 1.25x
2 (a) P
QD
QS
5
10
15
20
25
325
0
250
50
175
100
100
150
25
200
45 degree line
200
C 1 I line
150
The graphs are sketched in Figure S1.20.
100
P
50
25
20
50
15
100
150
200
Y
Figure S1.21
Y = 160
10
5
50
100
150
200
250
300
350
Q
Figure S1.20
If MPC decreases, the slope decreases as
indicated by the dashed line. The point of
intersection shifts down the 45° line, showing
that the equilibrium value of Y decreases.
7 (a) The coefficient of P2 in the demand equation for
good 1 is positive, indicating that demand for
good 1 rises as the price of good 2 goes up.
Similarly, for the second equation. Hence the
goods are substitutable.
(b) P1 = 15, P2 = 8; Q1 = 98, Q2 = 157
(b) P = 18, Q = 130
(c) The price of good 1 reduces by $3.15 and the
(c) P < 18
price of good 2 reduces by $0.35.
(d) The demand curve is unchanged and the supply
curve shifts upwards by 5 units.
P = 20, Q = 100
(b) slope = −1/3, vertical intercept = 16
(c) P = 3.6, Q = 37.2
(iii) Q5
(c) S5
4 (a) 1850
(b) 5
9 (a) x 5
5 (a) 12
36 2 5P
2P
2Y 2 100
Y12
3 2 2n
6n 2 m
;m=9
, y5
2 92m
2(9 2 m)
(i) Infinitely many solutions
(ii) No solutions
(b) 20
2QL
(c) K5
L2Q
C+I
(b) (i) 3
(ii) 5
3 (a) (1) is demand and (2) is supply.
6 Y
8 (a) 3f + 4s ≤ 3000; 640 kg
(b) P1 = 12, P2 = 7, P3 = 8
0
100
200
10 Y = 3160, r = 2.4
40
115
190
Figure S1.22 (not to scale) shows the IS and LM curves
intersecting at the equilibrium point (2.4, 3160). The
dashed line shows the new IS curve after a reduction
in MPC. The point of intersection has moved down
The lines are sketched in Figure S1.21.
674 ANSWERS
and to the left indicating a decrease in the equilibrium
values of both Y and r.
(b) x
f(x)
0
1
2
3
4
5
6
−9
−4
−1
0
−1
−4
−9
The graph is sketched in Figure S2.2.
Y
LM
y
1
2
3
4
5
6
3400
x
22
(2.4,3160)
24
26
2800
28
IS
210
f(x) 5 2x2 1 6x 2 9
212
34
r
Figure S2.2
Figure S1.22
(c) x
−2
−1
0
1
2
3
4
−12
−6
−4
−6
−12
−22
Chapter 2
f(x) −22
Section 2.1
The graph is sketched in Figure S2.3.
Practice Problems
y
1 (a) x = ±10
(b) x = ±2
(c) x = ±1.73
22
(d) x = ±2.39
(e) No solution
(f) No solution
3
4
x
210
2 (a) x = 10 and 2
1
2
(b) x 5 2
(c) No solution
3
2
215
220
(d) x = 2 and 3
−1
f(x)
21
0
1
5
2
−3
3
−3
The graph is sketched in Figure S2.1.
5
4
225
21
230
f(x) 5 22x2 1 4x 2 6
Figure S2.3
4 (a) The graph is sketched in Figure S2.4.
y
f(x) 5 4x2 2 12x 1 5
y
20
f(x) 5 2x2 2 11x 2 6
15
1
6
22
10
26
5
1
21
2
3
4
x
11
( 4 , 2169
)
8
25
Figure S2.1
2
25
(g) x = 0
3 (a) x
1
21
Figure S2.4
x
ANSWERS 675
(b) The graph is sketched in Figure S2.5.
5 The graphs are sketched in Figure S2.6.
y
y
y
(50, 2500)
f(x) 5 x2 2 6x 1 9
4
24
216
(a)
9
0
x
100
x
(b)
y
y
(11, 36)
81
5
(3, 0)
17
x
x
Figure S2.5
285
(c)
x
9
(d)
y
1
5 (a) 2 # x # 6
2
(b) The solution consists of all values of x except
3
for x = 3.
(b) x < −2, x ≥ 1
7 Q = 2 and P = 38
Exercise 2.1
1 (a) ±9
(b) ±6
(c) ±2
x
(e)
6 (a) 1 ≤ x ≤ 4
Figure S2.6
6 (a) x ≤ −4, x ≥ 4
(b) 0 < x < 100
(c) 5 ≤ x ≤ 17
(d) x = 9
(e) all values of x
7 (a) 56.166, 56.304, 56.35, 56.304, 56.166, 55.936,
(d) −2, 4
55.614, 55.2
(e) −9, −1
The graph is sketched in Figure S2.7.
2 (a) 1, −3
(b) 1/2, −10
(c) 0, −5
Q
57
(d) −5/3, 9/4
(e) 5/4, 5
56.5
3 (a) 0.44, 4.56
(b) −2.28, −0.22
56
(c) −0.26, 2.59
(d) −0.30, 3.30
55.5
(e) −2
(f) No solutions
4 (a) −4, 4
55
23
24
25
26
27
28
29
30 T
(b) 0, 100
(c) 5, 17
(d) 9
(e) No solution
Figure S2.7
(b) The production level is a maximum at 25°C, so
as temperature increases, output will fall.
676 ANSWERS
8 (a) x < 0, x > 3
(b) x ≤ −1, x ≥ 1
(c) −4 < x < 2
AC 5
9 Q = 4, P = 36
100
12
Q
The graph of the total cost function is sketched in
Figure S2.9.
10 P = 22, Q = 3
11 (a) $277
2 TC = 100 + 2Q
(b) 85
Exercise 2.1*
1 (a) ±13
TC
(b) −3, 13
2 −7d, d
3 (a) 3, −8
2
9
,2
3
2
(b)
(c) −2, 9
(c) 0,
slope 5 2
3
4
(d)
1
(twice)
6
(e) 2, −1, 4
intercept 5 100
4 (a) 7, 8
(b) 0.22, 2.28
(c) ±3
(d) 7 (twice)
(e) No solutions
(f) 10, 19
5 (a) x ≤ −8, x ≥ 8
(b) 1 ≤ x ≤ 9
1
2
(d) 21 # x #
(c) 27 , x , 2
Q
Figure S2.9
5
3
The graph of the average cost function is sketched in
Figure S2.10.
(e) x = −1
AC
6 c = 12; 6
12
7 k = 27
8 (a) x ≤ −3, x ≥ 4
(c) x ≤ 1, 2 ≤ x ≤ 3
(b) −1 < x < 2
10
(d) 2 ≤ x < 3, x > 5
8
9 $2250
6
10 120.76
4
11 (a) P = 18
(b) B = 15
AC 5 100 1 2
Q
2
12 (b) between $18 and $40 (c) $29
50
13 P = 5, Q = 65
Section 2.2
150
200 Q
Figure S2.10
3 π = −Q 2 + 18Q − 25
Practice Problems
1 TR = 1000Q − Q
100
The graph of the profit function is sketched in
Figure S2.11.
2
The graph is sketched in Figure S2.8.
Q = 500 and P = 500
(9, 56)
p
TR
p 5 2Q2 1 18Q 2 25
(500, 250 000)
31
4
TR 5 1000Q 2 Q2
1.52
225
0
Figure S2.8
1000
Q
Figure S2.11
14
16.48
Q
ANSWERS 677
(a) Q = 4 and 14
(b) Q = 9; π = 56
Exercise 2.2
3 (a) P = 50 − 4Q
(b) P 5
1 (a) P = 50; TR = 500
10
Q
4 TC = 500 + 10Q; AC 5
(b) TC = 150
(c) π = 350
500
1 10
Q
The graphs are sketched in Figures S2.15 and S2.16.
2 (a) 4Q
(b) 7
TC
(c) 10Q − 4Q2
The graphs are sketched in Figures S2.12, S2.13 and
S2.14.
slope 5 10
intercept 5 500
TR
TR 5 4Q
Q
Figure S2.15
Q
AC
Figure S2.12
AC 5 500 1 10
Q
TR
TR 5 7
10
7
Q
Q
Figure S2.16
Figure S2.13
TR
5 TC = Q2 + Q + 1; AC 5 Q 1 1 1
1
Q
The graphs are sketched in Figures S2.17 and S2.18.
(1.25, 6.25)
TR 5 10Q 2 4Q2
TC
0
2.5
1
Q
TC 5 Q2 1 Q 1 1
Q
Figure S2.14
Figure S2.17
678 ANSWERS
(c) π = TR − TC = 60Q − Q2 − (Q2 + 6Q + 100)
AC
= −2Q2 + 54Q − 100
AC 5 Q 1 1 1 Q1
e.g. multiply out brackets
Q = 2, 25
When Q = 13.5, π = 264.5.
Exercise 2.2*
the Q 1 1
line
2
3
1 15 and ; max profit is 154
1
1
5
which occurs at Q 5 7 .
12
6
2 a+b+c=9
Q
4a + 2b + c = 34
Figure S2.18
9a + 3b + c = 19
a = −20, b = 85, c = −56; π = −36
6 TC = 5Q + 100
3 (a) 20
7 (a) VC = $76
(b) $47 600
8 (a) $5.50
(b) x > 40
(b) 800
4 (a) TR 5
9 π = −2Q + 20Q − 32
2
(a) 2, 8
(b) 20
(b) TC = eQ + d
(c) 5
10 The graphs of TR and TC are sketched in
(c) AC 5 e 1
Figure S2.19.
(b) 3
(a) 1, 5
cQ 2 bQ2
a
(d) p 5
d
Q
2bQ2 1 (c 2 ae)Q 2 ad
a
5 (a) Ennerdale: 0.5aN + 0.25(1 − a)N
TR 5 22Q2 1 14Q
North Borsetshire: 15 + 0.3N
(3.5, 24.5)
maximum
profit
Equate two expressions to get result.
B
TC 5 2Q 1 10
L-shaped curve, tending to 1/5.
(b) a < 0.4 leads to N > 300, so choose Ennerdale if
number of withdrawals exceeds 300.
A
Section 2.3
Practice Problems
1 (a) 100
1
3
5
7
Q
(c) 1
(d) 1/10
(e) 1/100
(f) 1
(g) −1
(h) 1/343
(i) 81
( j) 72 101
(k) 1
2 (a) 4
Figure S2.19
11 (a) TR = PQ = 60Q − Q
(b) 10
2
The graph is an inverted parabola passing
through (0, 0), (60, 0).
(b) TC = 100 + (Q + 6)Q = Q2 + 6Q + 100
Divide by Q to find AC.
2
31, 26, 27 , 31; Q 5 10
3
3 (a) x6
(b) 3
(c) 32
(b) x1/2
(c) x6y12
(d)
1
4
(e) 1
(d) x3 + x1/2y 3
4 (a) f(K, L) = 7KL2
f(lK, lL) = 7(lK)(lL)2
= 7lK l2L2
(rule 4)
= (ll )(7KL )
2
2
= l3f(K, L)
Increasing returns to scale because 3 > 1.
(rule 1)
ANSWERS 679
13 (a) 98, 115, 125, 134, 140, 146
(b) f(K, L) = 50K1/4L3/4
(b) The graph is sketched in Figure S2.20.
f(lK, lL) = 50(lK) (lL)
1/4
3/4
= 50l K l L 1/4
(rule 4)
1/4 3/4 3/4
= (l l )(50K L )
1/4 3/4
1/4 3/4
N
= l1f(K, L)
(rule 1)
140
Constant returns to scale.
5 (1) (a) 3
(b) 2
(c) 1
(e) −1
(f) −2
(d) 0
120
(2) Same as part (1).
100
xz
y
80
6 (a) logb
60
(b) log b(x4y 2)
7 (a) x = 1.77 (b) x = 1
40
Exercise 2.3
1 (a) 64
(e) 1
11
(b) 2
(c) 1/3
(d) 1
(f) 6
(g) 4
(h) 1/343
5
(b) b
(c) c
(d) x 2y 2
(e) x 3y6
(f) y−4
(g) x4
(h) f 7
(i) y 3
( j) x5
2 (a) a
1
6
1
(b) x−2
3 (a) x 2
1
2
2
(e) x
(f) x
4 (a) 3600
(c) x 3
(d) x−1
3
2
5 The functions in parts (a) and (b) are homogeneous
of degree 7/12 and 2, respectively, so (a) displays
decreasing returns to scale and (b) displays increasing
returns to scale. The function in part (c) is not
homogeneous.
6 (a) 2 (b) −1 (c) −3 (d) 6
1
(c) 0
8 (a) log b(xz) (b) logb
(d) /2
4
6 t
5
10 (a) 1.29 (b) 1.70 (c) 6.03 (d) 8.31
1
4
1 (a) 8
(b) 1/32
(c) 625
(d) 2
2 (a) y2
(b) xy2
(c) x4y2
(d) 1
(c) x−3/2
(d) 2x11/2
(e) 2
4 3x3y7
x
y
(c) logb 3
2
y
z
1
11 (1) (a) 5 (b) 2
2
Exercise 2.3*
(e) −1
(c) 2 logb x + 7 logb y
(e) 2/3
2
(f) 5pq
(b) x1/4
(e) 8x−4/3
5 A[b(lK)a + (1 − b)(lL)a]1/a
(rule 4)
= A[blaK a + (1 − b)laLa]1/a
= A[(la)(bK a + (1 − b)La)]1/a
(factorise)
= A(la)1/a [bK a + (1 − b)La]1/a
(rule 4)
= lA[bK + (1 − b)L ]
(rule 3)
a 1/a
a
so f(lK, lL) = l f (K, L) as required.
1
2
x
y4
(3) 69.7
(2) (a) x 3y
3
The number of complaints increases but at a
decreasing rate.
3 (a) x−7
9 (a) 2 logb x + logb y (b) logb x − 2 logb y
12 (1) (a) 4
2
Figure S2.20
(e) 1/2 (f) 0
3
(2) logb
1
14 9
(b) 200 000
7 (a) 2 (b) 1
20
(b) −2
(c) 2
(b) x15y5
(c) x 2 y 2
6 (a) 2/3
(b) 3
7 (a) logb(1)
(b) logb
(c) 1/4
x3
y2
(d) logb(b2x3)
8 (a) 2 logb x + 3 logb y + 4 logb z
(b) 4 logb x − 2 logb y − 5 logb z
(c) logb x 2
1
1
logb y 2 logb z
2
2
(c) logb
x 5y
z2
680 ANSWERS
9 (a) −q
(b) 2p + q
10 (a) 78.31 (b) 1.48
(c) q − 4r
(d) p + q + 2r
(c) 3
(d) 0.23
y
27
11 (a) x ≤ 0.386 (3 dp)
24
(b) x > 14.425
21
y 5 32x
12 x = 3
18
2
3
15
13 (2) ; constant returns to scale.
12
2
1
14 (1) (a) (b) 2
3
2
(2) y 5
9
6
1 2
x
7
3
23
15 log10 x2 2 log10 y 2 log10 10
5 log10
x2
y
x4
y
5 log10
16 (a) L 5
3−x
2
3
to that of part (b).
1/b
Q
AK a
(b) L 5
(Q / A ) a 2 bK a
12b
3 (1) (a) 0.07%
1/a
(b) 1.35%
(c) 18.44%
(d) 50.06%
(2) 55%
(3) A graph of y against t, based on the information
obtained in parts (1) and (2), is sketched in
−3
−2
−1
0
1
2
3
0.04
0.11
0.33
1
3
9
27
27
9
3
1
0.33
0.11
0.04
The graphs of 3x and 3−x are sketched in Figures S2.21
and S2.22, respectively.
Figure S2.23. This shows that, after a slow start,
microwave ownership grows rapidly between
t = 10 and 30. However, the rate of growth then
decreases as the market approaches its saturation
level of 55%.
y
55
50
45
40
35
y
y5
30
27
25
24
20
y 5 3x
21
55
1 1 800e20.3t
15
18
10
15
5
12
5
9
10 15 20 25 30 35 40 45 50 t
6
Figure S2.23
3
23
22
Figure S2.21
x
(b) 2.718 281 828; values in part (a) are getting closer
21
Practice Problems
3
1
2 (a) 2.718 145 927, 2.718 268 237, 2.718 280 469
Section 2.4
x
21
Figure S2.22
21
17 (a) 1 (b) 2 (c) n (d) 3
1 x
22
21
1
2
3
x
4 (a) 2 ln a + 3 ln b
(b) ln
5 (a) $5 million and $3.7 million
(b) 4 years
x1/2
y3
ANSWERS 681
6 (1) Missing numbers are 0.99 and 2.80.
(2) The graph is sketched in Figure S2.24.
Intercept, 0.41; slope, 0.20.
(3) A = 0.2, B = e
0.41
(d) 5 ln x − 7 ln y (e) 1/2 ln x − 1/2 ln y
(f) 1/2 ln x + 3/2 ln y − 1/2 ln z
= 1.5
3 (a) ln x3 (b) ln
(4) (a) 9100
2 (a) ln x + ln y (b) ln x + 4 ln y (c) 2 ln x + 2 ln y
(b) 2.4 × 108; answer to part (b) is unreliable since
t = 60 is well outside the range of given data.
4 (a) 1.77
(d) 0.87
x 4z5
y3
(b) −0.80
(c) no solution
(e) 0.22
(f) 0.35
5 A = 50 000, a = 0.137 (a) $25 205 (b) $0
In s
6 (a) e5 (b) 1
3.0
7 91.6
2.5
8 Increasing returns to scale.
2.0
Exercise 2.4*
1.5
1 $162.19
1.0
2 (a) 1.13
(b) 1.79
(c) 8.77
3 The equation
0.5
N = c(1 − e−kt ) = c − ce−kt
1 2 3 4 5 6 7 8 9 10 11 12 13 t
rearranges as
e2kt 5
Figure S2.24
c2N
c
Taking logarithms gives
Exercise 2.4
1 (1) (a) 33
(b) 55
c2N
c
2kt 5 ln
(c) 98
(2) 100
(3) The graph of N against t is sketched in
Figure S2.25.
The graph sketched in Figure S2.25 is called a
learning curve. It shows that immediately after
training the worker can produce only a small
number of items. However, with practice, output
quickly increases and finally settles down at a
daily rate of 100 items.
c2N
1
t 5 2 ln
c
k
c2N
1
5 ln
c
k
21
5
1
c
ln
k
c2N
(a) 350 000
(b) 60 days
(c) Market saturation level is 700 000, which is
less than the three-quarters of a million copies
needed to make a profit, so the proprietor
should sell.
N
100
4 ln Q 5 ln 3 1
90
80
70
N 5 100 2 100e20.4t
60
50
1
1
ln L 1 ln K
2
3
Putting y = ln Q, x = ln K gives
y = 1/3x + (ln 3 + 1/2 ln L)
which is of the form ‘y = ax + b’.
40
Slope = 1/3, intercept = ln 3 + 1/2 ln L.
30
5 (a) ln Q = ln(ALn) = ln A + ln Ln = ln A + n ln L
20
10
(b) ln L
1
2
3
4
5
6
7
8
9 10 11 12 t
ln Q
0
0.69
1.10
1.39
1.61
−0.69
−0.46
−0.33
−0.22
−0.16
(c) n = 0.34, A = 0.50
Figure S2.25
682 ANSWERS
6 (a) 3y2 + 13y − 10
(b) Put y = ex in part (a) to deduce −0.405.
7 (a) y 5
x
1
ln
b
a
8 (a) 6
(b) −1 or 2
(b) x 5
1
ln(e y 2 3)
2
(c) 25
x6
y
5 (a) (i) loga
(ii) −2
(d) 0.26
(e) ±1.655
ln(A/B)
9 P5
k1 1 k2
(b) Each year the price is multiplied by the same
factor, 1.1; A = 1000; b = 1.1
(i) $2594
(ii) At the beginning of Year 22
Multiple Choice Questions
14
1 2. The graphs are
Q
sketched in Figures S2.27 and S2.28.
6 (a) TC = 14 + 2Q; AC =
1 B
2 D
3 C
4 B
5 C
6 D
7 C
8 A
9 A
10 B
11 D
12 E
13 D
14 B
15 A
TC
y
Examination Questions
1 (a) 2000
14
(b) 100
(c) 64
(d) Show f(lK, lL) = l7/6 f(K, L); increasing returns
to scale
2
π = −Q2 + 6Q − 8. A graph of the profit function is
Q
sketched in Figure S2.26
Figure S2.27
y
2
1
21
22
23
24
25
26
27
28
29
210
1
2
3
4
5
6
AC
x
Figure S2.26
2
Q
(a) Q = 2, 4
(b) 1
Figure S2.28
(c) Shifts vertically upwards
3 (a) P = 100, Q = 6
(b) P = 101.79, Q = 5.88; consumer pays $1.79 and
producer pays $3.21
4 (a) TR = 10Q − Q2
(b) 5 ≤ P ≤ 7
(b) When Q = 0, there is no revenue and the only
costs are the fixed ones which are given to be 14,
so the profit is 0 − 14 = −14. Substituting Q = 0
into π = aQ2 + bQ + c gives π = c, so c must
equal −14.
(c) a + b = 8, 6a + b = 3; a = −1, b = 9
ANSWERS 683
(d) Q = 2, 7; max profit is 6.25
7 (a) 5.7% increase
8 The complete set of ‘constant Year 2 prices’ is listed
5 3
ab
2
7 (a) (i) (b) 18.4% increase
(c) 12.0% increase
(e) TR = −Q + 11Q; P + Q = 11
2
in Table S3.1.
(ii) 4x2y4
During Year 1/2 salaries remain unchanged in real
terms. However, since Year 2 salaries have outpaced
inflation with steady increases in real terms.
(iii) 6
(iv) logb(xy2z2)
Table S3.1
(b) −1.36
(c) (i) 659,360
Year
(ii) 3.5
(iii) The sales graph is sketched in Figure S2.29.
Real salaries
1
2
3
4
5
18.1
18.1
19.0
21.7
23.2
Exercise 3.1
S
1 (a)
2 million
1
7
22
7
1
(b)
(c) 2 (d)
(e)
2
20
25
40
500
2 (a) 1.2
(b) 7.04
3 (a) 60%
(b) 22
4 (a) 1.19
(b) 3.5
5 (a) 4% increase
t
Figure 2.29
(c) 2190.24
(d) 62.72
(c) 0.98
(d) 0.57
(b) 42% increase
(c) 14% decrease
(d) 245% increase
(e) 0.25% increase
(f) 96% decrease
6 (a) $18.20
(b) 119 244
(d) $1610
(e) $15 640
(c) $101.09
7 35%
8 (a) A = 50,000; b = 0.95; A is the initial price
8 (a) $15.50
(b) $10.54
(c) 32%
(b) r = −0.0513
9 $862.50
(c) Use rules to logs to obtain ln V = −0.0513t + ln 50000
10 (a) $26 100
(b) 31% (nearest percentage)
11 (a) 37.5% increase
(b) 8.9% increase
so it is a line with gradient −0.0513 and vertical
intercept 10.8.
1
5
9 (a) 1 0.5 1 0.04Q; 1 0.5; 1
Q
Q
(b) A = IT; B = Printing; C = Retail
(c) 6.25% decrease
12 $11.6 million
13 50 60 72 86 100
Prices rise consistently over the past five years at an
steadily increasing rate.
(c) 2.5 < Q < 10
14 (a) Jan
Chapter 3
(b) 4800
(c) 133
15 (1) 1995
Section 3.1
(2) (a) 30%
(b) 52.3%
(c) 13.1%
(d) 9.4%
Practice Problems
(3) (a) 25%
(b) 44%
(c) 10.6%
(d) 11.1%
1 (a) $0.29
(b) $937.50
(c) $139.20
2 (a) 10%
(b) $1564
(c) $8835
3 (a) $7.345 million
(b) $76 billion
(c) 7%
4 (a) 8750
(b) 750
(c) 80%
5 (a) 82% increase
(b) 58% decrease
(c) 45% decrease
6 100, 101.5, 105.4, 104.3, 106.9
(4) Public transport costs have risen at a faster rate
than private transport throughout the period
1995–2010. However, for the final five years there
are signs that the trend has stopped and has
possibly reversed.
16 964, 100, 179, 750; e.g. seasonal variations.
17 (a) 83.3, 100, 91.7, 116.7, 125, 147.9
(b) 64.8
(c) Year 5
684 ANSWERS
Exercise 3.1*
4 13.86%
1 $140
5 12.55%
2 $5875
6 Nine years
3 (a) $90
Exercise 3.2
(b) 40%; the 20% discount is applied to the price
after the first reduction not the original.
4 (a) $720
(b) 40% of the new price is less than 40% of the
original.
2 $23 433.19
3 (a) $619 173.64
(b) 13
4 15 years
5 $42 868.75
(c) Divide by 0.6.
5 (a) $850
1 $6753.29; 50%
(b) 19% decrease
(c) 23.5% increase
6 (a) 100, 101.7, 113.1, 116.9
(b) Real output: 236, 229.2, 244.7, 244.7
6 (a) $13 947.94
(b) $14 156.59
(c) $14 342.45
(d) $14 381.03
7 Account B has the greater return.
8 $205.44
Index: 100, 97.1, 103.7, 103.7
(c) In real terms, spending on education fell by 2.9%
in Year 2, increased by 6.8% in Year 3 and
remained unchanged in Year 4.
7 (a) 1 and 6, respectively
9 36.6 years
10 17.0 years
11 We are charged interest on the interest; 26.82%.
12 7.25%
(b) 142, 150
13 (a) Six years (b) 5.19%
(c) 94, 87, 83, 75, 79
14 $4410; $5143.82; 28.60%
(d) 1.1 million and 1.6 million
15 21.70%
8 (a) 239.2, 242, 243.69, 243.43, 250.73, 258.56
98.8, 100, 100.7, 100.6, 103.6, 106.8
(b) 297
(c) 3.5%
Year 2 appear to cancel each other out, the firm
buys more of C, so this has the greater impact
on the index. In Year 3, there is a slight fall in the
price of good B. However, the firm buys relatively
few items of good B and the prices of the other
two goods have increased significantly.
10 (a) 100, 107, 146
(b) No change in Year 1 since this is the base year.
In Year 2 the firm has bought more of each good
so the index has increased.
In Year 3 the firm has nearly doubled its order for
good A which has also increased markedly in price.
This is reflected in a much higher index than before.
Section 3.2
Exercise 3.2*
1 13 years
2 $158.45
3 (a) Midwest (b) BFB
4 (a) $35 000
1 S = 1000(1.08)10 = $2158.92
(b) Seven years
5 7.67%
6 Store B
7 (a) Interest is (r/k)% per period and there are kt
periods in t years, so S 5 P 1 1
(b) If m 5
kt 5
Practice Problems
3 (1) (a) $33.71
(b) $33.77
(c) $33.79
(d) $33.81
(e) $33.82
(f) $33.82
r
100k
kt
.
100k
r
1
then
5 and
r
100k
m
mrt
so
100k
1
S 5 P 11
m
2 Four years
(2) $33.82
2n
18 (a) $28 000 (b) $25 920 (c) Year 5
(b) Although the price changes to goods A and C in
r
100
17 5.75%
9 (a) 100, 97, 112
16 P 5 S 1 1
rtm/100
by rule 3 of indices.
5P
1
11
m
m
rt/100
ANSWERS 685
(c) Now, since m = 100k/r, we see that if the frequency
increases (i.e. if k S ∞), then m S ∞, causing
11
1
m
m
to approach e. Substituting this into the result of
part (b) gives
S = Pert/100
1/n
10 100(g
rSn − Sn
= arn − a
(r − 1)Sn
= a(rn − 1)
(factorise both sides)
Sn 5 a
8 22 years
9 (a) 1.5%
have the extra term, arn. Consequently, when Sn is
subtracted from rSn, the rest of the terms cancel,
leaving
(b) 17.87%
− 1)
rn 2 1
r 21
(divide through by r − 1)
The expression for Sn denotes the sum of the first n
terms of a geometric series because the powers of r
run from 0 to n − 1, making n terms in total. Notice
that we are not allowed to divide by zero, so the last
step is not valid for r = 1.
11 (b) 100er/100 − 100
Section 3.3
Practice Problems
1 The geometric ratios of (a), (c), (d) and (e) are 2, −3,
8 (a) $480
1
/2 and 1.07, respectively.
(b) $3024.52
Sequence (b) is not a geometric progression.
(c) After n payments, debt is
2 (a) 16; 31
(b) 4386.52
((((8480 − A)R − A)R − A)R . . . − A)
3 (a) $15 645.49
(b) 12 years
= 8480Rn−1 − A(1 + R + R2 + . . . + Rn−1)
4 $177.69
5 8480Rn21 2 A
5 36 years
Finally, setting this expression equal to zero gives
the desired formula for A.
Exercise 3.3
1 11 463.88
2 (a) $78 227.44
(b) $78 941.10
3 $983.26
(d) $637.43
Section 3.4
4 17 years
Practice Problems
5 200 million tonnes
1 (a) $55 839.48
6 $19 053.06
(b) $54 881.16
7 (a) $13 586.80
(b) $35 868;
Rn 2 1
R21
$18 698.20
8 $966.43
3 NPV of Project A is $2221.90; NPV of Project B is
$2354.70, so Project B is to be preferred.
1 −16 777 215
(b) $9028.14
4 $180 146.91
5 The results are given in Table S3.2.
3 140 040
4 $424.19
(a) $459.03
(b) r = 16%; worthwhile since the IRR exceeds the
market rate.
Exercise 3.3*
2 (a) $9280.71
2 (a) NPV = $452; worthwhile since this is positive.
(b) $456.44
5 $313 238
6 $31 876.08
7 rSn = r(a + ar + ar 2 + . . . + ar n −1)
= ar + ar 2 + ar 3 + . . . + ar n
which is very similar to the given expression for Sn
except that the first term, a, is missing and we
There is very little to choose between these two
projects. Both present values are considerably
less than the original expenditure of $10 000.
Consequently, neither project is to be recommended,
since the net present values are negative. The firm
would be better off just investing the $10 000 at 15%
interest!
686 ANSWERS
Table S3.2
Exercise 3.4*
Discounted revenue
1 $257.85
2 $5000
($)
End of year
Project 1
Project 2
3 $31 250
1
2
3
4
5
Total
1739.13
1512.29
1972.55
1715.26
1491.53
8430.76
869.57
756.14
1315.03
3430.52
1988.71
8359.97
4 $61 672.67
5 $38 887.69
6 27%
7 $349.15
8 (a) $400 000
6 9%; this exceeds the market rate, so the project is
worthwhile.
7 The present values of the income stream are listed
in Table S3.3.
(b) $92 550.98
(c) $307 449.02
9 (a) $333 million
1 2 (1.06)12n
1.062 2 1.06
(b) 5
Table S3.3
Cash flow
Present value
End of year
($)
($)
1
2
3
Total present vlaue
70
70
1070
64.81
60.01
849.40
974.22
(c) Same expression as (b) but with 5 replaced
by 50.
(d) 12
10 $49 280
11 R(1 + r/100)−1 + R(1 + r/100)−2 + . . . + R(1 + r/100)−n
5
Exercise 3.4
R
1 2 (1 1 r/100)2n
1 1 r/100 1 2 (1 1 r/100)21
1 (a) $5974.43
5R
(b) $5965.01
1 2 (1 1 r/100)2n
1 1 r/100 2 1
2 (a) 7%
(b) Yes, provided there are no risks.
5 100 R
3 Project 2
4 $4 567 138.81
1 2 (1 1 r/100)2n
r
(a) $1488.94
5 Option 2
(b)
12 (b) 1%
6 6.27%
7 (a) The discount rate is certainly between 4% and
5%. An estimate of about 4.1% or 4.2% is
sensible.
(b) No
100R
r
Multiple Choice Questions
1 A
2 D
3 E
4 D
5 B
6 C
7 E
8 D
9 C
8 Project B is best.
10 C
11 E
12 D
9 (a) $379.08
13 C
14 E
15 E
(b) $1000
10 Project A
Project B
Choose B
11 20.3%
16 D
PV = 626.38
PV = 1248.28
Examination Questions
1 (a) $4880.76
(b) $4885.61
(c) 4.06%, 4.08%
(d) 2.94
ANSWERS 687
The graph of the cube function is sketched in
Figure S4.1.
2 (a) 2.62% decrease
(b) $607.15
f ′(−1) = 3.0; f ′(0) = 0; f ′(1) = 3.0
(c) $230,000
3 (a) (i) 3950
(ii) 3905
y
3
(c) Decreases but settles down to 3500
y 5 x3
4 (a) 122.54
(b) 135.21
2
(c) Paasche has the advantage of reflecting changes
in actual purchases made. However, the amounts
bought in Year 2 may be unavailable, and you are
not really comparing like with like in the Paasche
index.
1
5 4.77% and 4.88%. Very little to choose between
22
these two projects which are both close to the
prevailing rate. Other considerations such as project
viability and access arrangements need to be
considered in order to make an informed decision.
Dy 5 1.5
21
Dx 5 0.5
Dx 5 0.5
1
Dy 5 1.5
horizontal
tangent at
the origin
21
6 (a) 24
(b) (i) 44
(ii) $324.06
(iii) Remains at $4000
22
7 (a) (i) $4776.21
23
(ii) $2500
Figure S4.1
(iii) 2.8%
(iv) 5
3 f ′(x) = 3x2
Hence f ′(−1) = 3; f ′(0) = 0; f ′(1) = 3.
(b) (i) $40,554.48
(ii) $125,000
4 (a) 5x4
(b) 6x5
(d) −x−2
8 (a) $24 464.35
(b) $11 876.26
(c) 100x99
(e) −2x−3
Exercise 4.1
(b) −1
1 (a) 2
Chapter 4
(c) 0
2
2 − /3; downhill.
3 The graph of f(x) = 5 is sketched in Figure S4.2. The
Section 4.1
graph is horizontal, so it has zero slope at all values
of x.
Practice Problems
1 (a) 2
2 x
(b) −1
(c) 0
2 Using a calculator, the values of the cube function,
y
correct to two decimal places, are
y55
6
x
−1.50
−1.25
−1.00
−0.75
f(x)
−3.38
−1.95
−1.00
−0.42
x
−0.50
−0.25
0.00
0.25
0.50
f(x)
−0.13
−0.02
0.00
0.02
0.13
x
0.75
1.00
1.25
1.50
f(x)
0.42
1.00
1.95
3.38
5
4
3
2
1
25 24 23 22 21
Figure S4.2
1
2
3
4
5
x
688 ANSWERS
4 7x6; 448
4 (a) 14 (b) 6/x4 (c) 0
5 (a) 8x7
6 (a) 2
(b) 50x49
3
x4
(c) 19x18
1
(b)
(d) 999x998
1
(c) 2
(d)
2 x
2x x
7 3, 1.25, 0, −0.75, −1, −0.75, 0, 1.25
(a) −3
(b) 0
6 0; horizontal tangent, i.e. vertex of parabola must be
3 x
2
at x = 3.
7
(c) 1
When x = 1, y = a(1) + b = a + b.
1
3
x
(c)
3x
(d) 2
1
44 xx
13 x
(e) 2
2x8
(b) x−2/3 (c)
3 21/4
x
2
5
2
(d) 2 x23/2
(b) 50 − 6Q (c) −30/Q2
(f) −6Q2 + 30Q − 24
(e) 5/ L
3
2
2 (a) 4P + 1
(b) 40 2
(d) 4Y + 3
9
Q
2
(e) 200 2
1
4
L L
(c) 2
20
17
Q2
(f) −3Q2 + 40Q − 7
3 −40; concave
3 (a) (i) 2, 2.0248...
(ii) 0.24845...
(iii) 0.25
(b) (i) 8, 8.3018...
(ii) 3.01867...
(iii) 3
(c) (i) 0.5, 0.4938...
(ii) −0.06135... (iii) −0.0625
(b) (±3, ±243)
4 (a) (8, 4)
1
2 ,4
2
4,
(d)
4 (a) uphill; f ′(−1) = 1 > 0
(b) concave; f ″(−1) = −72 < 0
In all three cases the gradient of the chord gives a
good approximation to that of the tangent.
(c)
5
2 x
Exercise 4.2*
✓
(a 1 b) 2 b
5a
120
2 (a) 15x14
(a)
x
(d) 3
1 When x = 0, y = a(0) + b = b. ✓
9x3 x
(b)
2
1
8 (a) 2P + 1
Exercise 4.1*
Slope 5
5 4.
5 f ″(x) = 6ax + 2b > 0 gives x > −b/3a
f ″(x) = 6ax + 2b < 0 gives x < −b/3a
6 y=x−3
7 (a) f9(x) 5 2
1
8
(b) f 0(x) 5
Section 4.2
aA
, 0 so the function is decreasing.
xa11
a(a 1 1)A
. 0 so the function is convex.
x a12
(c) The graph is sketched in Figure S4.3.
Practice Problems
1 (a) 12x2
(b) −2/x2
2 (a) 5x 4 + 1
(b) 2x
3 (a) 2x − 3x2 (b)
3
x4
4 (a) 45x4 + 4x (b) 40x7 + 3/x2
(c) 2x + 6
(d) 8x3 + 36x2 − 8x + 7
5 134
Exercise 4.2
(b) −3/x2
1 (a) 10x
(d) 2x + 1
(c) 2
(e) 2x − 3
(f) 3 + 7/x
(h) a
(i) 2ax + b ( j) 4 1 3/x 2 2 14/x 3
2 (a) 27
2
(b) 4
(g) 6x − 12x + 49
2
(c) 2
(d) −36
(e) 3/8
(d) For salaries over $100 000, the proportion of
3 4x3 + 6x2
(a) 9x2 − 8x (b) 12x3 − 6x2 + 12x − 7 (c) 2x − 5
(d) 1 1
3
x2
Figure S4.3
(e) 2
2
4
1 2 3
x
x
(f)
3
10
2 3
2
x
x
people who earn a fixed range of salaries
decreases as income rises so, for example, there
are fewer people earning between $150 000 and
$200 000 than between $100 000 and $150 000.
8 r = γ which is a constant
ANSWERS 689
Section 4.3
MR
Practice Problems
100
1 TR = 60Q − Q2
MR 5 100 2 2Q
(1) MR = 60 − 2Q; −40
(2) (a) 500
(b) 459
50
so TR changes by − 41, which is approximately
the same as the exact value obtained in part (1).
100
Q
2 MR = 1000 − 8Q
2100
(a) Total revenue rises by about 2280.
(b) Total revenue falls by about 1520.
3 MC = 2, so a 1-unit increase in Q always leads to
a 2-unit increase in TC irrespective of the level of
output.
4 TC = 15 + 2Q2 + 9Q; 15; 4Q + 9
5 (a) 49.98
4 Q = 50L1/2
(a) 25
Figure S4.5
25
3
(b)
(b) 49.8
(c) 0.25
(c) 48
The fact that these values decrease as L increases
suggests that the law of diminishing marginal
productivity holds for this function.
(d) 30. Yes, because d2Q/dL2 = −0.02 < 0.
6 MPC = 1/6 and MPS = 5/6. If national income rises
by 1 unit, the approximate increase in consumption
and savings is 1/6 and 5/6, respectively.
5 MPS = 0.6; MPC = 0.4
This indicates that, at the current level of income,
a 1-unit increase in national income causes a rise
of about 0.6 units in savings and 0.4 units in
consumption.
7 13
8 At midday on 6 January, the company’s share price is
increasing at a rate of 25 cents a day.
By 7 January shares will have risen by 25 cents
(approximately).
Exercise 4.3
1 TR = 100Q − 4Q2, MR = 100 − 8Q; 1.2
9 TR = 3000Q − 2Q3/2;
2 TR = 80Q − 3Q , so MR = 2P − 80
2
MR = 2991; if Q rises from 9 to 10 the approximate
change in TR is 2991.
3 TR = 100Q − Q ; MR = 100 − 2Q. Graphs of TR
2
and MR are sketched in Figures S4.4 and S4.5,
respectively. MR = 0 when Q = 50. This is the value
of Q at which TR is a maximum.
MR 5 3000 2 3 Q
Exercise 4.3*
1 (a) TR = 100Q − 4Q3/2 − 3Q2
(b) MR = 100 − 6Q1/2 − 6Q; MR = 28
(c) 7 compared to 6.78
TR
2 (a) MPC = 0.96, MPS = 0.04
2
TR 5 100Q 2 Q
2500
(b) S = 0.2Y − 100 − 0.01Y2
3 (a) TC = 100 + 2Q + Q2/10
MC = 2 + Q/5
(b) MC = 8; Δ(TC) ≅ 16
(c) 100
4
0
Figure S4.4
50
100
Q
d 2Q
= 12 − 1.2L < 0 for all L > 10
dL2
5 (a) MPL 5
5 21/2
2 0.1
L
2
(b) L = 625; output is maximised when L = 625.
(c)
6 36
d 2Q
5
5 2 L23/2 , 0
4
dL 2
690 ANSWERS
7 MC = 20, so if Q increases by 1 unit, TC increases by
20 units.
MR = 18 at Q = 219, so if Q increases to 220 units,
Exercise 4.4*
1 (a) 20(2x + 1)9
(b) 3(x2 + 3x − 5)2(2x + 3)
TR rises by 18 approximately.
(c) −7/(7x − 3)2
The overall change in profit is approximately
18 − 20 = −2, so profit decreases.
(d) −2x/(x2 + 1)2
(e) 4/√(8x − 1)
8 25; when Q increases from 25 to 26, the approximate
increase in profit is 25.
9
2
(f) 22(6x 2 5)
(c)
Practice Problems
1 (a) 15(3x − 4)4
22
(2x 2 3)2
2 (a) (3x − 1)5(21x − 1)
(c)
x
(2x 1 3)
(b) 3(x2 + 3x + 5)2(2x + 3)
(d)
1 3x2 (2x 1 3) 5
2
(4x 2 3)
(f)
(i)
1
2x 1 1
21
x2(7x 1 9)
(2x 1 3)
5
(c)
3x 2(x 1 5)
(3x 1 5)3
24
(2x 1 1)3
100 2 3Q
200 1 50Q
(b)
(2 1 Q)3/2
(100 2 2Q)
7 1.098; −0.098; if income rises by 1 unit, consumption
6 (a)
(b) 16(2x − 7)7
(c) 5(x + 9)4
(e) 8(x + 2)(x2 + 4x − 3)
23
(g)
(3x 1 1)2
28
(h)
(4x 2 3)3
goes up by more than this, with the excess taken out
of savings.
8 Use the quotient rule. As Q increases, MC decreases
and converges to 2.
9 Use the product rule to differentiate
(2x 1 3) (2x 1 3)
(b) x(5x − 4)(x − 2)
2
3x 1 4
(d) (4x + 3)(x + 6)2
2 x12
(e) (8x + 13)(x + 5)2
(f) x2(14x − 15)(2x − 5)3
25
7
25
3 (a)
(b)
(c)
(x 2 5)2
(x 1 7)2
(x 2 2)2
(d)
3(x 2 1)
(g) (6x2 + 17x + 6)(x + 2)(x + 3)2
2 (a) (9x + 4)(3x + 4)
(c)
x 2(5x 2 6)
3
(c)
2
(x 1 1)
2(x 2 1)3/2
2x 2 3
1 2 x2
ad 2 bc
(d) 2
(e)
(cx 1 d)2
(x 1 1)2
(f) [ac(m + n)x + mad + ncb](ax + b)m−1(cx + d)n−1
Exercise 4.4
(d) 24x(4x2 − 7)2
(b)
4 (a) (x − 3)3(5x − 3) (b)
22
2
(b)
(x 2 2)2
(x 1 1)2
1 (a) 15(5x + 1)2
x3(9x 1 8)
2 (x 1 1)
x(x 1 8)
3 (a)
(x 1 4)2
22
(c)
(x 2 2)2
3 (a)
22
(b) x4(4x + 5)(28x + 25)
Section 4.4
(b)
5
(6x 2 5) 3 (6x 2 5)
2 (a) 5x(x + 2)(x + 5)2
d(AC)
c
MC 2 AC
5a2 2 5
dQ
Q
Q
3
4
3
225
6
219
(e)
(f)
(3x 1 1)2
(5x 1 6)2
(3x 2 7)2
4 10(5x + 7) = 50x + 70
Section 4.5
Practice Problems
1 − 0.26
2 (a) If P = 10 then |E| = 1/9 < 1, so inelastic.
(b) If P = 50 then |E| = 1, so unit elastic.
(c) If P = 90 then |E| = 9, so elastic.
3 2
5 7x6 + 24x5 + 20x4
6 (a) (100 − 4Q)(100 − Q)2 (b)
TR = aQ − Q(bQ + c)1/2.
4000
(Q 1 4)2
7 MPC = 1.78, MPS = −0.78. If national income rises
by 1 unit, consumption rises by 1.78 units, whereas
savings actually fall by 0.78 units.
47
; 7.7%
36
4 (a) 0.333175
(b)
1
3
Exercise 4.5
1 −43/162 = −0.27
2 −22/81 = −0.27; agree to two decimal places.
3 (a) −1/4
(b) −1/4
(c) −9/8
ANSWERS 691
4 (a) −0.3125, so inelastic.
(b) Would expect demand for economy class flights
7 (a) Q 5
to be more sensitive to price rises, so the
magnitude of E would be larger.
(b) a 5
6 (a) 1 2
400
(b) 200
Q
(c) MR = 20 − 0.1Q
7 (a) 0.2P
2
,b59
5
8 (b) (i) 3
(ii) 1.5
5 P
(iii)
10 P 2 4
Section 4.6
(b) 0.1P 2 = Q − 4
Practice Problems
(subtract 4 from both sides)
1 (a) There is one minimum point at (−2, −47).
P = 10(Q − 4) = 10Q − 40
2
(multiply both sides by 10)
A graph is sketched in Figure S4.6.
(b) Minimum at (2, −1) and maximum at (3, 0).
P = √(10Q − 40)
A graph is sketched in Figure S4.7 based on the
following table of function values:
(square root both sides)
dP
5
5
dQ
(10Q 2 40)
(10Q 2 40)
1
(c)
5
5
dP/dQ
5
dQ
1
5
dP
a
a
1
P
3 5
P2b
a
P2b
E5P3
5 (a) −2, so when price rises by 2%, demand falls by 4%.
(b) −3.88%
P2b
a
x
−10
0
2
3
10
f(x)
3887
27
−1
0
−833
P
dQ
5 0.2P 5
5
dP
y
y 5 3x 2 1 12x 2 35
(d) E = 10/7
8 1.46;
(a) elastic
(b) 7.3%
x
Exercise 4.5*
1 −1.54
235
2 3.75%
3 4P/(4P − 60); 7.5
4 E52
5
(22, 247)
1
which is a constant.
n
Figure S4.6
aP
aP 1 b
(a) If b = 0, then E 5
Pa
5 1.
Pa
(b) If b < 0, then aP + b < aP, so E 5
y
aP
. 1.
aP 1 b
27
Assuming that the line is sketched with quantity
on the horizontal axis and price on the vertical
axis, supply is unit elastic when the graph passes
through the origin, and elastic when the vertical
intercept is negative.
(3, 0)
x
(2, 21)
6 (1) 0.528
(2)
0.2P 2
(a) 4.2% (b) 20
40 1 0.1P 2
y 5 22x 3 1 15x 2 2 36x 1 27
Figure S4.7
692 ANSWERS
(c) Minimum at (10, 5); graph is sketched in
2 L = 10; MPL = APL = 2000
Figure S4.10.
3 (a) Q = 5
(b) Q = 4 gives the maximum profit, π = 30;
MR = MC = 4
y
4 Q = 6; AC = MC5
t = 12.5
105
Exercise 4.6
1 (a) Maximum at (1/2, 5/4); graph is sketched in
Figure S4.8.
y
1
2
y 5 x 2 2 20x 1 105
5
4
( , )
y 5 2x 2 1 x 1 1
1
(10, 5)
x
x
Figure S4.10
(d) Maximum at (1, 2), minimum at (−1, −2); graph
is sketched in Figure S4.11.
Figure S4.8
y
(b) Minimum at (2, 0); graph is sketched in
Figure S4.9.
y 5 2x 3 1 3x
(1, 2)
y
2 3
4
3
x
(21, 22)
y 5 x2 2 4x 1 4
(2, 0)
x
Figure S4.11
Figure S4.9
2 10
3 30; MPL = 450 = APL
4 (a) TC = 13 + (Q + 2)Q
= 13 + Q2 + 2Q
AC 5
13
TC
1Q12
5
Q
Q
Q
1
2
3
4
5
AC
16
10.5
9.3
9.3
9.6
6
10.2
The graph of AC is sketched in Figure S4.12.
(b) From Figure S4.12 minimum average cost is 9.2.
(c) Minimum at Q = √13, which gives AC = 9.21.
ANSWERS 693
AC
16
(a)
y
15
14
AC 5
13
13
Q
1Q12
x
12
y5x
11
3
10
9
8
(b)
7
1
2
3
4
5
y
6 Q
y 5 x4
Figure S4.12
Q2
4
5 (a) TR 5 4Q 2
3
p5
x
2
Q
2Q
1 2Q 2 4
1
20
20
MR 5 4 2
(c)
Q
2
y
x
3Q 3Q 2
MC 5 2 2
1
5
20
(b) 4 (c) MR = 2 = MC
y 5 2x 6
6 $3
7 (a) TC = 2Q2 − 36Q + 200
AC 5
200
TC
5 2Q 2 36 1
Q
Q
(b) Q = 10
Figure S4.13
1 AC = 4
3 TC 5 2Q 2 1 15, AC 5 2Q 1
2
At Q 5 10,
(c) Q = 10
d (AC)
5 0.4 . 0
dQ2
1
min.
MC = 4 = AC
8 56.25
9 $440
10 37 037
Exercise 4.6*
1 531.5
2 Graphs of the three functions are sketched in
Figure S4.13, which shows that the stationary points
in (a), (b) and (c) are a point of inflection, minimum
and maximum, respectively.
15
,
Q
MC = 4Q; √7.5; AC = 11 = MC
4 (a) (6, 492) is a maximum and (10, 460) is a
minimum.
(b) The graph is sketched in Figure S4.14.
694 ANSWERS
Section 4.7
S
600
Practice Problems
1 (a) TR = (25 − 0.5Q)Q = 25Q − 0.5Q2
500
TC = 7 + (Q + 1)Q = Q2 + Q + 7
MR = 25 − Q
400
MC = 2Q + 1
(b) From Figure S4.16, the point of intersection of
300
the MR and MC curves occurs at Q = 8. The
MC curve cuts the MR curve from below, so this
must be a maximum point.
200
100
2
4
6
8
10
12
25
t
Figure S4.14
MC
(8, 17)
(c) 492 and 465
MR
5 167
1
6 (a) 15
(b) Q 5 15 2
25
t
2t
gives P 5 140 1
10
5
P rises from 140 to 140 1
2t
2
, so the increase is t.
5
5
7 a = −7, b = 16, c = −7
Q
Figure S4.16
2 MC = 100
(a) P1 = $200 and P2 = $150
8 (a) The graph is sketched in Figure S4.15.
(b) P = $500/3
With discrimination, profit is $10 000.
Without discrimination, profit is $8333.
40
3 Domestic market: |E| = 2
35
Foreign market: |E| = 3
30
We see that the firm charges the higher price in the
domestic market, which has the lower value of |E|.
25
20
Exercise 4.7*
15
1 (a) TR = aQ2 + bQ, TC = dQ + c
10
(b) MR = 2aQ + b, MC = d
5
(c) The equation 2aQ + b = d has solution
1
2
3
4
5
Figure S4.15
(b) TR = 39.0625, Q = 6.875
(c) 5
6
7
8
9
10
Q5
d2b
2a
2 (a) At the point of maximum total revenue
d(TR)
50
dQ
so E = −1
MR 5
(b) Maximum occurs when Q = 10.
ANSWERS 695
3 (a) P1 = $30, P2 = $20
(b) P = $24.44
The profits in parts (a) and (b) are $95 and
$83.89, respectively.
The graph of the natural logarithm function is
sketched in Figure S4.17.
f 9(1) 5 1
4 Argument is similar to that given in text but with
< replaced by >.
d2C
5 2DRQ23 . 0
dQ2
5 (a)
(b) 2DRH
6 (a) EOQ = 40 with C = $4000
f 9(2) 5
1
2
f 9(3) 5
1
3
These results suggest that f′(x) = 1/x.
(b) EOQ = 50 with C = $3200
(c) EOQ = 80 with C = $8000
y
(d) Reducing (increasing) the order costs by a factor
of k reduces (increases) the minimum total cost
by a factor of k . The same applies to changes
in the holding costs.
7 AC 5 aQ 1 b 1
Dx 5 0.5
Dy 5 0.15
Dx 5 0.5
Dy 5 0.25
1
Dx 5 0.5
Dy 5 0.5
c
; 2 ac 1 b
Q
1
0
8 The argument is similar to that given in the text
3
4 x
y 5 ln x
21
for APL.
2
9 The new supply equation is
P = aQS + b + t
Figure S4.17
In equilibrium
which has solution
Q5
2
d2b2t
a1c
4 (a)
(d) 1/x
x 1 2 2 x ln x
x(x12)2
7x 1 6
2x 1 6
(b)
x(x 1 2)
x(2x 1 3)
5 In terms of P the total revenue function is given by
(d 2 b)t 2 t 2
tQ 5
a1c
TR = PQ = 1000Pe−0.2P
and the total cost function is
which differentiates to give
TC = 100 + 2Q = 100 + 2000e−0.2P
d 2 b 2 2t
a1c
Hence
This is zero when
π = TR − TC
d2b
t5
2
= 1000Pe−0.2P − 2000e−0.2P − 100
For maximum profit, P = 7.
Also the second derivative is
22
, 0 (since a and c are both positive)
a1c
which confirms that the stationary point is a
maximum.
6 E = −2.05
Exercise 4.8
1 (a) 6e6x
(b) −342e−342x
−x
(d) 40e4x − 4x
(c) −2e
+ 4e x
2 (1) (a) $4885.61
Section 4.8
0.04t
Practice Problems
f(x)
(c) 1/x
3 (a) x3(1 + 4 ln x) (b) 2xex (c)
Hence
1 x
(b) −e−x
2 (a) 3e3x
aQ + b + t = −cQ + d
0.50
1.00
1.50
2.00
−0.69
0.00
0.41
0.69
x
2.50
3.00
3.50
4.00
f(x)
0.92
1.10
1.25
1.39
(2) 160e
; 195.42
1
x
(b)
1
x
(b)
4x3 1 6x
x4 1 3x2
3 (a)
4 (a) 3x2ex
3
(b) $4887.57;
196
696 ANSWERS
5 (a) (4x3 + 2x4)e2x
Exercise 4.8*
(b) 1 + ln x
6 (a)
(b)
1 (a) 2e2x + 12e−4x (b) (4x + 1)e4x (c)
2e4x(2 x2 2 x 1 4)
(x2 1 2)2
(d) (m ln x + 1)x m−1
ex(x ln x 2 1)
x(ln x)2
(g)
Figure S4.18.
Figure S4.19.
y
(n ln x 2 1)x n21
(ln x)2
(amx 1 bm 2 an)emx
(ax ln bx 2 n)eax
(h)
(ax 1 b)n11
x(ln bx)n11
x
2e
(i)
(ex 1 1)2
7 (a) Maximum at (1, e−1); the graph is sketched in
(b) Maximum at (1, −1); the graph is sketched in
(e) ln x (f)
2(x 1 2)e]x
x3
2 (a)
1
1
9x 2 2
(b)
(c)
1 2 x2
x(1 1 x)
2x(3x 2 1)
3 (a)
dy
4 y 5 Akekt 4 Aekt 5 k
dx
(b) (i) a and b
e21
GDP
A
5 e(a 2 b)t which is exponential with a
N
B
growth rate of (a − b).
2
12 x
4 (a) 4x 3(1 2 x 2)e22x (b)
x(x 1 1)
(ii)
1
x
y 5 xe2x
1 2e2 1
.
a
a
5 (a) Maximum at 2 ,
(b) Maximum at 2
Figure S4.18
b
b2
, ln 2
2a
4a
.
4(x 1 1)
(2x 1 1)(4x 1 3)
7 100
6 (b)
y
1
8
x
(2Q 1 1)(3 ln(2Q 1 1) 2 20)
6Q
9 (a)
21
(2Q2 2 1)eQ
Q2
10 (a) E 5
y 5 lnx 2 x
2
(b)
1
2Q
1 ln
3Q 1 1
3Q 1 1
2P
so demand is inelastic when 0 ≤ P < 100.
100
(b) 100
11 2
12 (a)
Figure S4.19
8 (a) 2359.88, 2313.15 so the decrease is 46.7.
dS
(b)
5 247.2 so is a good approximation.
dt
9 49
10 50
11 E 5 2
(b)
dy
abke2at
5
. 0 so the gradient is positive.
dt
(1 1 be2at )2
d2y
a2bke2at (21 1 be2at )
5
dt 2
(1 1 be2at)3
All of the factors are positive for all values of t
except for the factor −1 + be−at which is positive
when t ,
ln b
ln b
and negative when t .
.
a
a
k
and
11b
approaches k as t S ∞ because e−at S 0.
(c) The curve crosses the y axis at 0,
10
, which is −1 when Q = 10.
Q
ANSWERS 697
(d) The graph is sketched in Figure S4.20.
y
40
y
k
20
24
2
22
4
x
220
240
k
11 b
Figure S4.21
t
(ii) −1/3
(iii) 1
Figure S4.20
13 (a) PV 5 Ve20.1t 5 2e
t
3 e20.1t
(b) 25
Multiple Choice Questions
2 A
3 B
4 A
5 A
6 E
7 C
8 A
9 E
10 D
11 A
12 A
13 B
14 C
15 B
16 D
17 E
18 B
19 B
20 E
21 C
22 D
23 A
24 C
Examination Questions
35 Q
2 ;
2
2
(a) An increase of 0.625
1 MR =
4 1225
2 (b) −6x + 10x + 24; 3, 36 , 2 , 2
3
27
4
(c) −12x + 10; max at x = 3 and min at x = 2
3
(d) The graph is sketched in Figure S4.21. It crosses
the x-axis at −3, 1 and 4.5, and it crosses the
y-axis at −27.
2
(b) (i) 25/41 = 0.610
(ii) 30/49 = 0.612
4 (a) (i) 31, 34.222; 32.22
1 E
(b) 35
3 (a) (i) 90
(ii) 31
(iii) Gradients of chord and tangent very similar
with the chord slightly steeper
(b) (i) MC = 6Q + 8
(ii) 94 versus 94.75 so the % error in the estimate
is 0.79%
5 (a) 7
(b) 7/3
(c) 6/7%, elastic
6 (a) (i) 12e2x
2x
x 11
(b) 14, 16;
(ii) 2
(i) increasing
(ii) convex
7 (i) 48
(ii) 294
(iii) 1/2
(iv) 3/8
(v) 0; output maximised at L = 50
8 (a) π(0) = −16 < 0; π(1) = 1 − 9 + 24 − 16 = 0
(b) (2,4), (4,0)
(c) Max and min, respectively
698 ANSWERS
(d) The graph of the profit function is sketched in
Figure S4.22.
p
5
P5
Lines intersect only if the vertical intercept on the
supply graph is above that on the demand graph.
d 2c
ad 1 bc
ad 1 bc 1 ct
d2b2
(b) P5
, Q5
; P5
, Q5
a 1c
a1c
a1c
a1c
ad 1 bc
d 2c
ad 1 bc 1 ct
d2b2t
, Q5
; P5
, Q5
a 1c
a1c
a1c
a1c
1
2
3
4
5
t
(c) c:a
25
(d) t5
210
d 2 b d2T
22
; 2 5
, 0, so max
2
dt
a1c
12 (a) 0
215
(b) 20; AC minimum at Q = 20
220
(c) 40; −31/12
13 (a) E 5
Figure S4.22
2P
; solve E = −1 to get P = a/3
P2a
(b) TR = aQ + bQ3/2; Q 5
(e) increasing on 0 ≤ t < 2, t > 4 and decreasing on
2<x<4
so max
(f) t < 3, t > 3
14 (a) Bookwork
3
2
(g) After an initial loss of 16, the profit increases so
that it breaks even after one year and reaches a
local maximum of 4 after two years. The profit
then declines so that the firm breaks even again
after four years. For the next year the profit rises
at an increasing rate, reaching 4 at the end of the
five-year period.
9 (a) 10 + 12L − 0.3L2; 10 + 6L − 0.1L2
(b) P1 = 60, P2 = 135; E1 5 2 , E2 5 2
27
;
23
Bookwork
15 Bookwork
16 (a) 32/55; 32/11; inelastic
(b) a = −4, b = 80
Chapter 5
(b) 30
Section 5.1
2
(c) d Q . 0 for L > 20
2
Practice Problems
dL
400 2 9Q
10 (a) 400 2 6Q
1 (a) −10 (b) −1 (c) 2 (d) 21 (e) 0 (f) 21
20Y 1 Y 2 2 200
(b) (10 1 Y )2
11 (a) The supply and demand curves are sketched in
Figure S4.23.
2 (a)
(b)
−f
−f
5 0 2 2y 5 22y
5 20x 3 2 0 5 20x 3;
−x
−y
−f
−f
5 3x2y2 2 0 5 3x2y2
5 2xy3 2 10;
−y
−x
3 (a) fxx = 60x2
fyy = −2
fyx = fxy = 0
P
P 5 aQ 1 b
(b) fxx = 2y3
d
fyy = 6x2y
P 5 2cQ 1 d
fyx = fxy = 6xy2
4
b
Q
Figure S4.23
3b
4a d2 (TR)
;
, 0,
5
4 Q
9b2 dQ2
f1 5
−f
5 x2 1 5x 41
−x1
f11 5
−2 f
5 20x 31
−x12
f21 5
−2 f
51
−x2−x1
ANSWERS 699
5
−z
−z
5 1,
54
−x
−y
Exercise 5.1*
(a) z increases by approximately 0.3.
2 f(kw, kx, ky) = 5(kw)0.34(kx)0.25(ky)0.41
(b) At (2, 6), z = 14, and at (1.9, 6.1), z = 14.29, so
= 5k0.34w0.34k0.25x0.25k0.41y0.41
= 5k0.34+0.25+0.41w0.34x0.25y0.41
= 5kf(w, x, y)
1 85 ≠ 91; (0, y) for any y.
the exact increase is 0.29.
6 (a)
dy
2y
5
dx x 2 3y2 1 1
2
(b)
dy
y
5 4
dx
5y 2 2xy
Exercise 5.1
1 324; 75; 0
2 (a) f (a, a) = 2a2 + aa = 3a2
(b) f (b, −b) = 2b2 + b(−b) = b2
3
f (2x, 2y) = (2x)(2y)2 + 4(2x)3
= 2x × 4y2 + 4 × 8x3
= 8xy2 + 32x3
= 8(xy2 + 4x3)
= 8f (x, y)
4 (a) fx = 2x, fy = 20y4
(b) fx = 9x2, fy = −2e y
3
fx
(a) y
(b) exy
(c) 2x + 2
(d) 4x−3/4y3/4
(e)
22y
x3
0
exy
2
0
0
0
1
ex
0
1
ex
0
x 6y
1 1
2 2 4
y x2
y x
2x
3
y
22
x3
2
1 22
1
2 2
y2 x3
y
we2vw,
−z
5 2uw we2vw,
−v
−z
u 2vw
e 2 uv we2vw
5
−w
2 w
fxy(e, 1) = −6
fx(1, 0) = −2
7 1/3
fy(1, 1) = 7
6 (a) −0.6
8
f1 5
(b) −2
(c) −2.6
x x2
1
2x x
x32
1
; f2 5 2 1 3 3 1 ; f3 5 1 3 1 ;
x2
x2
x2
x3
x2
f11 5 0; f22 5
7 (a) fx = −3x2 + 2, fy = 1
dy 2(23x2 1 2)
5 3x2 2 2
5
1
dx
(b) y = x3 − 2x + 1, so
✓
0z
0z
0z
5 1 1 2v,
5 2v 1 2u,
5 215w2
0u
0v
0w
0z
5 3u21/2v1/3w1/6, −z 5 2u1/2v22/3w1/6,
0u
−v
−z
5 u1/2v1/3w25/6
−w
9 (a) 21.5
(b) 24.575; 12.5%
x
ex
1
6 fx(e, 1) = 3; fy(e, 1) = −2e; fxx(e, 1) = 6e−1; fyy(e, 1) = 4e;
fy = 5x4y4 + 2y
(b)
fxy
12x1/4y−1/4 −3x−7/4y3/4 −3x1/4y−5/4 3x−3/4y−1/4 3x−3/4y−1/4
−z
5
−u
(b)
5 fx = 4x y − 2x
8 (a)
fyx
−z
5 12vw2(6u 1 vw3)3
−w
3 5
5 3x2 2 2
fyy
−z
−z
5 4w3(6u 1 vw3)3,
5 24(6u 1 vw3)3,
−u
−v
5 (a)
(d) fx = 6x5y2, fy = 2x6y + 15y2
dy
fxx
4 78; 94; 6.2
(c) fx = y, fy = x + 6
dx
1
fy
2x1x32
1
2x
1
2 2 ; f33 5 1 2 2 ;
x2
x23
x2
x3
f12 5
2x32
2x
5 f21; f13 5 3 5 f31;
x2
x22
f23 5
22x1x3
5 f32
x22
9 e.g. f(x, y) = x3y2 + 3x2y
10 5
11 2.5
12 (a)
(c)
23x2y 2 4y2
x3 1 8xy
(b)
24x22/3y1/4 2 1
3x1/3y23/4
2y 2
1 1 xy
(d)
2x 2 2 2xy 1 y 2
2x 2 1 2xy 1 y 2
Section 5.2
Practice Problems
1 (a) −0.14
(b) −0.14
(c) 0.12
0.6%. A rise in income causes a rise in demand. Normal.
700 ANSWERS
2
−U
−U
5 2948 and
5 140
−x1
−x2
ΔU = −848
four positive numbers.
The law of diminishing marginal utility holds for
both x1 and x2.
Utility increases as more units of good 1 are
consumed.
3 21.06;
$42.12
(a) MRTS 5
(b) K
−2U
5 0.35 Ax120.3 x220.5 . 0 since it is the
−x1−x2
(b)
and MPL = 4L
4 MPK = 2K
product of four positive numbers.
MPL
4L
2L
5
5
MPK
2K
K
Consuming more of one good increases the
marginal utility of the other good.
−Q
−Q
1L
5 K(2K) 1 L(4L)
−K
−L
✓
three positive numbers and one negative number.
Consuming more of good 1 decreases
the marginal utility of good 1; there is
diminishing marginal utility of good 1.
Exercise 5.2
1 2.71. This is greater than 1, so the good is superior.
2 (a) −4/233
(b) −3/233
6 The graph is sketched in Figure S5.1, which shows
(c) 40/233; −0.04%; complementary
that MRTS = −(−5/7) = 5/7.
3 (a) 2x2 + y2 = 36
(b) 16 and 4 so the gradient is −4.
4 (a) 80
−2U
5 20.21Ax121.3 x20.5 , 0 since it is the product of
−x12
(c)
5 2(K 2 1 2L2 ) 5 2Q
K
(b) 960
slope 5 2
100
52
140
5
7
120
5 1
6
−U
5 0.7Ax120.3 x20.5 . 0 since it is the product of
−x1
5 (a)
1
5
−U
−U
and
5
5
5
12
−x1
−x2
(a) 37/60
80
(b) 12/25
60
7 MPK = 8, MPL = 141/4 (a) 125/32 (b) 125/32
100
8 K(6K + 3L ) + L(6LK) = 6K + 9L K = 3(2K + 3L K)
40
9 (a) −0.5; −1.5 so A is inelastic and B is elastic, so B
20
2
2
140
100
3
2
3
2
is more sensitive to a change in price.
(b) 0.5; 1.5
20
A is normal because income elasticity is positive
and is less than one.
Figure S5.1
B is superior because income elasticity of
demand exceeds one.
7 (a) MPK 5
10 (a) 18(lK)1/6(lL)5/6 = 18l1/6K1/6l5/6L5/6 = l(18K1/6L5/6) = lQ
(b) MPK = 3K−5/6L5/6; MPL = 15K1/6L−1/6
40
60
10
;
3
80
MPL 5
1
21
2
1
21
MRTS 5
12b
La21 1 2 b K
MPL
5
3 a21 5
MPK
b
b
L
K
1
21
1
1 LA(1 2 b)La21[ ] a
1
4 A(lK) (lL) = Al K l L = l
a
b
a
a b b
L
MPK 5 AbK a21[ ] a ; MPL 5 A(1 2 b)La21[ ] a
(b) KAbK a21[ ] a
3 16
160
decrease in capital maintains a constant level
of output.
8 (a)
2 0.5
140
(c) 3.15; e.g. a 1-unit increase in labour and a 3.15
Exercise 5.2*
Since E < 1, the good is income-inelastic. The relative
market share of the good decreases as the economy
expands.
120
(b) ΔQ ≈ −11
(c) (i) decreases (ii) increases
1 (a) 0.4 (b) 3.2%
100
5 A(bK a 1 (1 2 b)La )[ ] a
Q
a+b
1
5 A[ ] a
5Q
21
21
21
12a
ANSWERS 701
9 (a) Complementary; coefficient of PA is negative.
An increase in X* leads to an increase in Y. An
increase in m leads to a decrease in Y.
2b
(b) (i) EP 5
100
(ii) EP 5 2
A
3c
500
(c) Y 5 2100; DY 5
3 If d increases by a small amount, then the intercept
increases and the demand curve shifts upwards
slightly. Figure S5.2 shows that the effect is to
increase the equilibrium quantity from Q1 to Q2,
confirming that ∂Q/∂d > 0.
d
(iii) EY 5
5
(c) a = 4310, b = 5, c = 2, d = 1
10 (a)
−Q
5 kbP2a PAb21Y c . 0; as the price of the
−PA
alternative good increases, demand for
the good increases.
P
higher
demand
curve
d 1 Dd
(c) The partial derivative of this proportion with
respect to Y is k(c − 1)P1−aPAbYc−2 > 0.
11 (a) MRCS 5
100
3
d
2
; constant so the indifference map
3
same
supply
curve
consists of straight lines.
(b) MRCS 5
3x 2
; decreases as x1 increases so the
x1
indifference curve is convex.
1
(c) MRCS 5
3
b
x2
; decreases as x1 increases so the
x1
Q1 Q2
indifference curve is convex.
Section 5.3
Q
Figure S5.2
Practice Problems
1 C5a
b 1 I*
1b
I2a
Exercise 5.3*
1 (a) Substituting equations (3) and (4) into (2) gives
C = a(Y − T*) + b = aY − aT* + b
−C
a
.0
5
−I* 1 2 a
Substituting (5), (6) and (7) into (1) gives
because 0 < a < 1
Y = aY − aT* + b + I* + G*
Hence an increase in I* leads to an increase in C.
so that
Change in C is 2.
Y5
2 (a) Substitute C, I, G, X and M into the Y equation
to get
2aT* 1 b 1 I* 1 G*
12a
Finally, from (7), we see that
Y = aY + b + I* + G* + X* − (mY + M*)
C5a
Collecting like terms gives
(1 − a + m)Y = b + I* + G* + X* − M*
so
Y5
(b)
(7)
b 1 I* 1 G* 1 X* 2 M*
12 a 1 m
−Y
1
5
−X*
12 a1 m
b 1 I* 1 G* 1 X* 2 M*
−Y
52
−m
(1 2 a 1 m)2
(b)
2
2aT* 1 b 1 I* 1 G*
2 aT* 1 b
12a
5
a(2aT* 1 b 1 I* 1 G*) 1 (1 2 a)(2aT* 1 b)
12a
5
aI* 1 aG* 2 aT* 1 b
12a
a
. 0; C increases (c) 1520; rise of 18
12a
2a(b 1 I* 1 G* 2 aT*
(1 2 a 2 at)2
702 ANSWERS
3 (1) From the relations
6 (1) Substituting second and third equations into first
gives
C 5 aYd 1 b
Y = aY + b + cr + d
Yd 5 Y 2 T
so that
T 5 tY 1 T*
(1 − a)Y − cr = b + d
we see that
C = a(Y − tY − T *) + b
(2) Substituting first and second equations into third
gives
Similarly,
k1Y 1 k2r 1 k3 5 MS*
M = m(Y − tY − T *) + M*
so that
Substitute these together with I, G and X into the
Y equation to get the desired result.
(2) (a)
−Y
m2a
5
−T* 1 2 a 1 at 1 m 2 mt
Numerator is negative because m < a.
Denominator can be written as
k1Y 1 k2r 5 M*S 2 k3(2)
(3) (a) Working out c × (2) + k2 × (1) eliminates r to
give
ck1Y 1 k2 (1 2 a)Y 5 c(M*S 2 k3) 1 k2 (b 1 d )
(1 − a) + at + m(1 − t)
which represents the sum of three positive
numbers, so is positive. Hence the autonomous
taxation multiplier is negative.
(b)
−Y
1
.0
5
−G* 1 2 a 1 at 1 m 2 mt
(b)
C 5 aYd 1 b 5 a[(1 2 t)Y 2 T*] 1 b 5 a(1 2 t)Y 2 aT* 1 b
Y 5 C 1 I 5 a(1 2 t)Y 2 aT* 1 b 1 cr 1 d
(1 2 a(1 2 t))Y 5 2aT* 1 b 1 cr 1 d 1
(c) ΔT* = 181/3
Y5
4 From text, equilibrium quantity is
(b)
Substituting this into either the supply or demand
equation gives the desired result.
0 , a(1 2 t) , 1 1 1 2 a(12t) . 0
r.01
1
(b 2 aT* 1 I* 1 G*)
12a
1
−Y
a
;
;
52
−G* 1 2 a −T*
12a
12a
DY 5
51
12a
(b) −Y 5
(c) 1
−Y
. 0 1 Y increases
−c
(d) Y = 2800; ΔY = 46.25
8 (a) π = (a − bQ)Q − ( f + vQ) = −bQ2 + (a − v)Q − f
At the stationary point, Q 5
Y 5 C 1 I 1 G 5 a(Y 2 T*) 1 b 1 I* 1 G*
Y5
−Y
r
;
5
−c 1 2 a(1 2 t)
(c) 0 , t , 1 and 0 , a , 1 1
5 (a) C 5 aYd 1 b 5 a(Y 2 T ) 1 b 5 a(Y 2 T*) 1 b
(1 2 a)Y 5 2aT* 1 b 1 I* 1 G* 1
b 1 d 2 aT* 1 cr
1 2 a(1 2 t)
2T* 1 (1 2 t)(b 1 d) 1 cr)
−Y
5
−a
(1 2 a(1 2 t))2
−P a(b 2 d)
−P
a
, 0,
.0
5
5
−c
−d
a1c
(a 1 c)2
where the quotient rule is used to obtain ∂P/∂a and
∂P/∂c. An increase in a, b or d leads to an increase in
P, whereas an increase in c leads to a decrease in P.
c
, which is positive because the top
(1 2 a)k2 1 ck1
7 (a) Yd 5 Y 2 T 5 Y 2 (tY 5 T*) 5 (1 2 t)Y 2 T*
(b) ΔY = 20
−P c(d 2 b)
−P
c
. 0,
.0
5
5
−a
−b
a1c
(a 1 c)2
Dividing both sides by ck1 + k2(1 − a) gives
result.
and bottom of this fraction are both negative.
(3) (a) 1000
d2b
a1c
(1)
(b)
a2v
.
2b
(a 2 v)2 −p
−p
a 2 v −p
,
,
52
5 21,
5
4b2
−f
−a
2b −b
−p
a2v
52
−v
2b
(c) From (a), Q 5
a2v
so a − v > 0.
2b
Hence all of the multipliers in (b) are negative
except for the first one.
An increase in a causes maximum profit to
rise, whereas an increase in b, f or v results
in a decrease in maximum profit.
ANSWERS 703
Section 5.4
(b) Q1 = 14; Q2 = 9; π = 358
Practice Problems
−2p
−Q12
1 One saddle point at (0, 1).
TR1 = P1Q1 = (50 − Q1)Q1 = 50Q1 − Q
2
1
Also
Total revenue from the sale of G2 is
TR2 5 P2Q2 5 (95 2 3Q2)Q2
−2p
−2p
, 0 so the point is a maximum.
2 , 0,
−Q1
−Q22
P5
104 2
2 Q.
3
3
5 50Q1 2 Q12 1 95Q2 2 3Q 22
p5
Profit is
104
2
1
2 Q Q 2 4Q 5 (92Q 2 2Q2)
3
3
3
p 5 TR 2 TC
5 (50Q1 2 Q21 1 95Q2 2 3Q22 ) 2 (Q12 1 3Q1Q2 1 Q22 )
(b) Q = 23; new p 5 352
1
3
3 P1 = 200, P2 = 150 and π = 10 000
−z
5 4x 2 12,
−x
1
2
−z
5 2y 2 8; (3, 4)
−y
−z
−2z
2
2
−x−y
−y
2
5 4 3 2 2 02 . 
Download